You are on page 1of 33

Clinical Infectious Diseases

IDSA GUIDELINE

Official American Thoracic Society/Infectious Diseases


Society of America/Centers for Disease Control and
Prevention Clinical Practice Guidelines: Diagnosis of
Tuberculosis in Adults and Children
David M. Lewinsohn,1,a Michael K. Leonard,2,a Philip A. LoBue,3,a David L. Cohn,4 Charles L. Daley,5 Ed Desmond,6 Joseph Keane,7
Deborah A. Lewinsohn,1 Ann M. Loeffler,8 Gerald H. Mazurek,3 Richard J. O’Brien,9 Madhukar Pai,10 Luca Richeldi,11 Max Salfinger,12 Thomas M. Shinnick,3
Timothy R. Sterling,13 David M. Warshauer,14 and Gail L. Woods15
1
Oregon Health & Science University, Portland, Oregon, 2Emory University School of Medicine and 3Centers for Disease Control and Prevention, Atlanta, Georgia, 4Denver Public Health Department,
Denver, Colorado, 5National Jewish Health and the University of Colorado Denver, and 6California Department of Public Health, Richmond; 7St James’s Hospital, Dublin, Ireland; 8Francis J. Curry
International TB Center, San Francisco, California; 9Foundation for Innovative New Diagnostics, Geneva, Switzerland; 10McGill University and McGill International TB Centre, Montreal, Canada;
11
University of Southampton, United Kingdom; 12National Jewish Health, Denver, Colorado, 13Vanderbilt University School of Medicine, Vanderbilt Institute for Global Health, Nashville, Tennessee,
14
Wisconsin State Laboratory of Hygiene, Madison, and 15University of Arkansas for Medical Sciences, Little Rock

Background.  Individuals infected with Mycobacterium tuberculosis (Mtb) may develop symptoms and signs of disease (tuber-

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


culosis disease) or may have no clinical evidence of disease (latent tuberculosis infection [LTBI]). Tuberculosis disease is a leading
cause of infectious disease morbidity and mortality worldwide, yet many questions related to its diagnosis remain.
Methods.  A task force supported by the American Thoracic Society, Centers for Disease Control and Prevention, and Infectious
Diseases Society of America searched, selected, and synthesized relevant evidence. The evidence was then used as the basis for rec-
ommendations about the diagnosis of tuberculosis disease and LTBI in adults and children. The recommendations were formulated,
written, and graded using the Grading, Recommendations, Assessment, Development and Evaluation (GRADE) approach.
Results.  Twenty-three evidence-based recommendations about diagnostic testing for latent tuberculosis infection, pulmonary
tuberculosis, and extrapulmonary tuberculosis are provided. Six of the recommendations are strong, whereas the remaining 17 are
conditional.
Conclusions.  These guidelines are not intended to impose a standard of care. They provide the basis for rational decisions in
the diagnosis of tuberculosis in the context of the existing evidence. No guidelines can take into account all of the often compelling
unique individual clinical circumstances.

EXECUTIVE SUMMARY Testing for LTBI


Individuals infected with Mycobacterium tuberculosis (Mtb) Our recommendations for diagnostic testing for LTBI are based
may develop symptoms and signs of disease (TB disease) or upon the likelihood of infection with Mtb and the likelihood of
may have no clinical evidence of disease (latent tuberculosis progression to TB disease if infected, as illustrated in Figure 1.
infection [LTBI]). TB disease is a leading cause of infectious • We recommend performing an interferon-γ release assay
disease morbidity and mortality worldwide, with many diag- (IGRA) rather than a tuberculin skin test (TST) in individuals
nostic uncertainties. A  task force supported by the supported 5 years or older who meet the following criteria: (1) are likely
by the American Thoracic Society, Centers for Disease Control to be infected with Mtb, (2) have a low or intermediate risk
and Prevention, and Infectious Diseases Society of America of disease progression, (3) it has been decided that testing for
appraised the evidence and derived the following recommen- LTBI is warranted, and (4) either have a history of BCG vacci-
dations using the Grading, Recommendations, Assessment, nation or are unlikely to return to have their TST read (strong
Development, and Evaluation (GRADE) approach (Table 1): recommendation, moderate-quality evidence). Remarks: A TST
is an acceptable alternative, especially in situations where an
Received 4 October 2016; editorial decision 6 October 2016; accepted 14 October 2016.
IGRA is not available, too costly, or too burdensome.
These guidelines were endorsed by the European Respiratory Society on 20 June 2016. • We suggest performing an IGRA rather than a TST in all
a
Authors are co-chairs of this guideline committee. other individuals 5  years or older who are likely to be
Correspondence: D. M. Lewinsohn, Pulmonary and Critical Care Medicine, Oregon Health &
Science University, Portland, OR (lewinsod@ohsu.edu). infected with Mtb, who have a low or intermediate risk
Clinical Infectious Diseases® 2017;64(2):e1–e33 of disease progression, and in whom it has been decided
© The Author 2016. Published by Oxford University Press for the Infectious Diseases Society of
that testing for LTBI is warranted (conditional recommen-
America. All rights reserved. For permissions, e-mail journals.permissions@oup.com.
DOI: 10.1093/cid/ciw694 dation, moderate-quality evidence). Remarks: A TST is an

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e1


Table  1. Interpretation of Strong and Weak (Conditional) (conditional recommendation, very low-quality evidence).
Recommendations
Remarks: In situations in which an IGRA is deemed the
preferred diagnostic test, some experts are willing to use
Weak (Conditional)
Strong Recommendation Recommendation IGRAs in children over 3 years of age.
Patients Most individuals in this situation The majority of individuals in • The preceding recommendations are summarized in Figure
would want the recommended this situation would want 2. While both IGRA and TST testing provide evidence for
course of action, and only a the suggested course of
small proportion would not. action, but many would
infection with Mtb, they cannot distinguish active from
not. latent TB. Therefore, the diagnosis of active TB must be
Clinicians Most individuals should receive Recognize that different excluded prior to embarking on treatment for LTBI. This
the intervention. Adherence to choices will be appropri-
this recommendation according ate for individual patients is typically done by determining whether or not symptoms
to the guideline could be used and that you must help suggestive of TB disease are present, performing a chest
as a quality criterion or perfor- each patient arrive at a
mance indicator. Formal decision management decision
radiograph and, if radiographic signs of active TB (eg, air-
aids are not likely to be needed consistent with his or her space opacities, pleural effusions, cavities, or changes on
to help individuals make deci- values and preferences.
serial radiographs) are seen, then sampling is performed
sions consistent with their Decision aids may be
values and preferences. useful in helping individ- and the patient managed accordingly.
uals to make decisions
consistent with their
values and preferences.
Policy The recommendation can be Policymaking will require
Testing for TB Disease

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


makers adopted as policy in most substantial debate and
situations. involvement of various • We recommend that acid-fast bacilli (AFB) smear micros-
stakeholders. copy be performed, rather than no AFB smear microscopy,
in all patients suspected of having pulmonary TB (strong
recommendation, moderate-quality evidence). Remarks:
acceptable alternative, especially in situations where an False-negative results are sufficiently common that a
IGRA is not available, too costly, or too burdensome. negative AFB smear result does not exclude pulmonary
• There are insufficient data to recommend a preference for TB. Similarly, false-positive results are sufficiently com-
either a TST or an IGRA as the first-line diagnostic test in mon that a positive AFB smear result does not confirm
individuals 5 years or older who are likely to be infected pulmonary TB. Testing of 3 specimens is considered the
with Mtb, who have a high risk of progression to disease, normative practice in the United States and is strongly
and in whom it has been determined that diagnostic test- recommended by the Centers for Disease Control and
ing for LTBI is warranted. Prevention and the National Tuberculosis Controllers
• Guidelines recommend that persons at low risk for Mtb Association in order to improve sensitivity given the
infection and disease progression NOT be tested for Mtb pervasive issue of poor sample quality. Providers should
infection. We concur with this recommendation. However, request a sputum volume of at least 3 mL, but the optimal
we also recognize that such testing may be obliged by law volume is 5–10  mL. Concentrated respiratory specimens
or credentialing bodies. If diagnostic testing for LTBI is and fluorescence microscopy are preferred.
performed in individuals who are unlikely to be infected • We suggest that both liquid and solid mycobacterial cultures
with Mtb despite guidelines to the contrary: be performed, rather than either culture method alone,
for every specimen obtained from an individual with sus-
• We suggest performing an IGRA instead of a TST in
pected TB disease (conditional recommendation, low-qual-
indivduals 5 years or older (conditional recommendation,
ity evidence). Remarks: The conditional qualifier applies to
low-quality evidence). Remarks: A TST is an acceptable
performance of both liquid and solid culture methods on
alternative in settings where an IGRA is unavailable, too
all specimens. At least liquid culture should be done on all
costly, or too burdensome.
specimens as culture is the gold standard microbiologic test
• We suggest a second diagnostic test if the initial test
for the diagnosis of TB disease. The isolate recovered should
is positive in individuals 5 years or older (conditional
be identified according to the Clinical and Laboratory
recommendation, very low-quality evidence). Remarks:
Standards Institute guidelines and the American Society for
The confirmatory test may be either an IGRA or a
Microbiology Manual of Clinical Microbiology.
TST. When such testing is performed, the person is
• We suggest performing a diagnostic nucleic acid amplifi-
considered infected only if both tests are positive.
cation test (NAAT), rather than not performing a NAAT,
• We suggest performing a TST rather than an IGRA in on the initial respiratory specimen from patients suspected
healthy children <5  years of age for whom it has been of having pulmonary TB (conditional recommendation,
decided that diagnostic testing for LTBI is warranted low-quality evidence). Remarks: In AFB smear-positive

e2 • CID 2017:64 (15 January)  •  Lewinsohn et al


Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017
Figure 1.  Paradigm for evaluation of those with latent tuberculosis infection (LTBI) based on risk of infection, risk of progression to tuberculosis, and benefit of therapy. In
developing a diagnostic approach for the evaluation of those with suspected LTBI, we recommend the clinician weigh the likelihood of infection, the likelihood of progression to
tuberculosis if infected, and the benefit of therapy (Horsburgh and Rubin, Clinical practice: latent tuberculosis infection in the United States. N Engl J Med 2011; 364:1441–8).
Recommendations were formulated for each of the 3 groups illustrated above. These groups are concordant with current recommendations for the interpretation of the tuber-
culin skin test (American Thoracic Society, Targeted tuberculin testing and treatment of latent tuberculosis infection. MMWR Recomm Rep 2000; 49:1–51). Abbreviations: CXR,
chest radiograph; HIV, human immunodeficiency virus; LTBI, latent tuberculosis infection; Mtb, Mycobacterium tuberculosis; RR, ; TB, tuberculosis; TST, tuberculin skin test.

patients, a negative NAAT makes TB disease unlikely. recommendation, moderate-quality evidence). Remarks:
In AFB smear-negative patients with an intermediate to This recommendation specifically addresses patients who
high level of suspicion for disease, a positive NAAT can are Hologic Amplified MTD positive because the Hologic
be used as presumptive evidence of TB disease, but a Amplified MTD NAAT only detects TB and not drug
negative NAAT cannot be used to exclude pulmonary resistance; it is not applicable to patients who are positive
TB. Appropriate NAAT include the Hologic Amplified for types of NAAT that detect drug resistance, including
Mycobacteria Tuberculosis Direct (MTD) test (San many line probe assays and Cepheid Xpert MTB/RIF.
Diego, California) and the Cepheid Xpert MTB/Rif test • We suggest mycobacterial culture of respiratory speci-
(Sunnyvale, California). mens for all children suspected of having pulmonary TB
• We recommend performing rapid molecular drug sus- (conditional recommendation, moderate-quality evidence).
ceptibility testing for rifampin with or without isoniazid Remarks: In a low incidence setting like the United States,
using the respiratory specimens of persons who are either it is unlikely that a child identified during a recent con-
AFB smear positive or Hologic Amplified MTD positive tract investigation of a close adult/adolescent contact
and who meet one of the following criteria: (1) have been with contagious TB was, in fact, infected by a different
treated for tuberculosis in the past, (2) were born in or individual with a strain with a different susceptibility
have lived for at least 1  year in a foreign country with pattern. Therefore, under some circumstances, microbi-
at least a moderate tuberculosis incidence (≥20 per 100 ological confirmation may not be necessary for children
000)  or a high primary multidrug-resistant tuberculosis with uncomplicated pulmonary TB identified through a
prevalence (≥2%), (3) are contacts of patients with multid- recent contact investigation if the source case has drug-­
rug-resistant tuberculosis, or (4) are HIV infected (strong susceptible TB.

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e3


Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017
Figure 2.  Summary of recommendations for testing for latent tuberculosis infection (LTBI). 1Performing a second diagnostic test when the initial test is negative is a
strategy to increase sensitivity. This may reduce specificity, but the panel decided that this is an acceptable trade-off in situations in which the consequences of missing LTBI
(ie, not treating individuals who may benefit from therapy) exceed the consequences of inappropriate therapy (ie, hepatotoxicity). 2Performing a confirmatory test following
an initial positive result is based upon both the evidence that false-positive results are common among individuals who are unlikely to be infected with Mycobacterium
tuberculosis and the committee’s presumption that performing a second test on those patients whose initial test was positive will help identify initial false-positive results.
Abbreviations: IGRA, interferon-γ release assay; LTBI, latent tuberculosis infection; TST, tuberculin skin test.

• We suggest sputum induction rather than flexible broncho- sampling whose induced sputum is AFB smear micros-
scopic sampling as the initial respiratory sampling method copy negative or from whom a respiratory sample cannot
for adults with suspected pulmonary TB who are either be obtained via induced sputum (conditional recommenda-
unable to expectorate sputum or whose expectorated spu- tion, very low-quality evidence). Remarks: Bronchoscopic
tum is AFB smear microscopy negative (conditional recom- sampling in patients with suspected miliary TB should
mendation, low-quality evidence). include bronchial brushings and/or transbronchial biopsy,
• We suggest flexible bronchoscopic sampling, rather than as the yield from washings is substantially less and the yield
no bronchoscopic sampling, in adults with suspected from BAL unknown. For patients in whom it is important
pulmonary TB from whom a respiratory sample can- to provide a rapid presumptive diagnosis of tuberculosis
not be obtained via induced sputum (conditional recom- (ie, those who are too sick to wait for culture results), trans-
mendation, very low-quality evidence). Remarks: In the bronchial biopsies are both necessary and appropriate.
committee members’ clinical practices, bronchoalveolar • We suggest that cell counts and chemistries be performed
lavage (BAL) plus brushings alone are performed for most on amenable fluid specimens collected from sites of sus-
patients; however, for patients in whom a rapid diagnosis pected extrapulmonary TB (conditional recommendation,
is essential, transbronchial biopsy is also performed. very low-quality evidence). Remarks: Specimens that are
• We suggest that postbronchoscopy sputum specimens be amenable to cell counts and chemistries include pleural,
collected from all adults with suspected pulmonary TB cerebrospinal, ascitic, and joint fluids.
who undergo bronchoscopy (conditional recommendation, • We suggest that adenosine deaminase levels be measured,
low-quality evidence). Remarks: Postbronchoscopy spu- rather than not measured, on fluid collected from patients
tum specimens are used to perform AFB smear micros- with suspected pleural TB, TB meningitis, peritoneal
copy and mycobacterial cultures. TB, or pericardial TB (conditional recommendation, low-­
• We suggest flexible bronchoscopic sampling, rather than quality evidence).
no bronchoscopic sampling, in adults with suspected mil- • We suggest that free IFN-γ levels be measured, rather
iary TB and no alternative lesions that are accessible for than not measured, on fluid collected from patients with

e4 • CID 2017:64 (15 January)  •  Lewinsohn et al


suspected pleural TB or peritoneal TB (conditional recom- These clinical practice guidelines on the diagnosis and classi-
mendation, low-quality evidence). fication of tuberculosis in adults and children were prepared by
• We suggest that AFB smear microscopy be performed, rather a task force supported by the American Thoracic Society (ATS),
than not performed, on specimens collected from sites of the Centers for Disease Control and Prevention (CDC), and the
suspected extrapulmonary TB (conditional recommenda- Infectious Diseases Society of America (IDSA). Additionally,
tion, very low-quality evidence). Remarks: A positive result Fellows of the American Academy of Pediatrics participated in
can be used as evidence of extrapulmonary TB and guide the development of these guidelines. The specific objectives of
decision making because false-positive results are unlikely. these guidelines are as follows:
However, a negative result may not be used to exclude TB
• To define high- and low-risk patient populations based
because false-negative results are exceedingly common.
upon the results of epidemiological studies.
• We recommend that mycobacterial cultures be performed,
• To provide diagnostic recommendations that lead to bene-
rather than not performed, on specimens collected from
ficial treatments and favorable clinical outcomes.
sites of suspected extrapulmonary TB (strong recommen-
• To describe a classification scheme for tuberculosis that is
dation, low-quality evidence). Remarks: A  positive result
based on pathogenesis.
can be used as evidence of extrapulmonary TB and guide
decision making because false-positive results are unlikely. These guidelines target clinicians in high-resource coun-
However, a negative result may not be used to exclude TB tries with a low incidence of TB disease and LTBI, such as the
because false-negative results are exceedingly common. United States. The recommendations may be less applicable to

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


• We suggest that NAAT be performed, rather than not per- medium- and high-tuberculosis incidence countries. For such
formed, on specimens collected from sites of suspected countries, guidance documents published by the World Health
extrapulmonary TB (conditional recommendation, very Organization (WHO) may be more suitable.
low-quality evidence). Remarks: A  positive NAAT result
HOW TO USE THESE GUIDELINES
can be used as evidence of extrapulmonary TB and guide
decision making because false-positive results are unlikely. These guidelines are not intended to impose a standard of care.
However, a negative NAAT result may not be used to exclude They provide the basis for rational decisions in the diagnos-
TB because false-negative results are exceedingly common. tic evaluation of patients with possible LTBI or TB. Clinicians,
At present, NAAT testing on specimens other than sputum patients, third-party payers, stakeholders, or the courts should
is an off-label use of the test. never view the recommendations contained in these guidelines
• We suggest that histological examination be performed, as dictates. Guidelines cannot take into account all of the often
rather than not performed, on specimens collected from compelling unique individual clinical circumstances. Therefore,
sites of suspected extrapulmonary TB (conditional rec- no one charged with evaluating clinicians’ actions should
ommendation, very low-quality evidence). Remarks: Both attempt to apply the recommendations from these guidelines by
positive and negative results should be interpreted in rote or in a blanket fashion. Qualifying remarks accompanying
the context of the clinical scenario because neither false-­ each recommendation are its integral parts and serve to facili-
positive nor false-negative results are rare. tate more accurate interpretation. They should never be omit-
• We recommend one culture isolate from each mycobac- ted when quoting or translating recommendations from these
terial culture-positive patient be submitted to a regional guidelines.
genotyping laboratory for genotyping (strong recommen-
dation, very low-quality evidence). METHODS

Persons infected with Mycobacterium tuberculosis (Mtb) have a Committee Selection


broad array of presentations, ranging from those with clinical, The criteria for committee selection were an (1) established track
radiographic, and microbiological evidence of tuberculosis (TB record in the relevant clinical or research area; (2) involvement
disease) to those who are infected with Mtb but have no clini- with the ATS Assembly on Microbiology, Tuberculosis and
cal evidence of TB disease (latent tuberculosis infection [LTBI]). Pulmonary Infections, the IDSA Tuberculosis Committee, or
Individuals with LTBI who have been recently exposed have employment by the United States CDC Division of Tuberculosis
an increased risk of developing TB, whereas those with remote Elimination; and (3) absence of disqualifying conflicts of interest.
exposure have less risk over time unless they develop a condi- Conflicts of interest were managed according to the policies and
tion that impairs immunity. Operationally, recent exposure can procedures agreed upon by the participating organizations [1].
be defined either epidemiologically (ie, as might occur in the set- The committee was divided into subcommittees assigned to
ting of the household of an infectious case or occupational expo- develop drafts for each of the following areas: (1) LTBI, (2) clini-
sure) or immunologically (ie, conversion of a tuberculin skin test cal and radiological aspects of TB diagnosis, (3) microbiological
or interferon-γ release assay [IGRA] from negative to positive). evaluation for TB diagnosis and detection of drug resistance,

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e5


and (4) pediatric TB diagnosis. Meetings were held either and undesirable consequences (ie, the benefits may not be worth
in-person or via teleconference. the costs or burdens), the balance of desirable and undesirable
consequences depends upon the clinical context, or there is
Evidence Synthesis variation about how individuals value the outcomes. A  strong
Each subcommittee identified key diagnostic questions and then recommendation should be interpreted as the right thing to do
performed a pragmatic evidence synthesis for each question, to for the vast majority of patients; a weak recommendation should
identify and summarize the related evidence. The subcommit- be interpreted as being the right thing to do for the majority of
tees first sought studies comparing one diagnostic intervention patients, but maybe not for a sizeable minority of patients.
with another and measuring clinical outcomes. Such evidence
was unavailable, so the subcommittees next sought diagnostic TUBERCULOSIS: EPIDEMIOLOGY, TRANSMISSION,
accuracy studies. When published evidence was lacking, the AND PATHOGENESIS
collective clinical experience of the committee was used. The
A full discussion of these topics can be found in the
evidence syntheses were used to inform the recommendations.
Supplementary Materials. TB disease remains one of the
Though comprehensive, the evidence syntheses should not be
major causes of morbidity and mortality in the world.
considered systematic reviews of the evidence.
The WHO estimates that 8.6 million new cases of tuberculosis
occurred in 2014 and approximately 1.5 million persons died from
Developing and Grading Recommendations the disease [4]. The emergence of drug-resistant tuberculosis has
Recommendations were formulated and the quality of evidence become apparent over the past 2 decades, and in particular, mul-

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


and strength of each recommendation were rated using the tidrug-resistant tuberculosis (MDR-TB; resistant to isoniazid and
Grading, Recommendations, Assessment, Development and rifampin) and extensively drug-resistant tuberculosis (XDR-TB;
Evaluation (GRADE) approach [2, 3]. resistant to isoniazid and rifampin, plus any fluoroquinolone and at
The quality of evidence is the extent to which one can be con- least 1 of 3 injectable second-line drugs [ie, amikacin, kanamycin,
fident that the estimated effects are close to the actual effects or capreomycin]), which are more difficult to treat than drug-sus-
and was rated as high, moderate, low, or very low. The quality of ceptible disease [5, 6]. The approximate number of cases of
evidence rating derived from the quality of the accuracy studies MDR-TB in the world is roughly 500 000 reported from at least 127
that informed the panel’s judgments, as randomized trials and countries, and XDR-TB has been reported from 105 countries [4].
controlled observational studies were lacking. Well-done accu- In the United States, 9412 cases of TB disease were reported in
racy studies that enrolled consecutive patients with legitimate 2014, with a rate of 3.0 cases per 100 000 persons. Sixty-six per-
diagnostic uncertainty and used appropriate reference stand- cent of cases were in foreign-born persons; the rate of disease was
ards represented high-quality evidence; lack of these charac- 13.4 times higher in foreign-born persons than in US- born indi-
teristics constituted reasons to downgrade the quality evidence. viduals (15.3 vs 1.1 per 100 000, respectively) [7]. An estimated
Normally, the quality of evidence for first-line therapy would 11 million persons are infected with Mtb [8]. Thus, although
have been factored into such quality of evidence ratings but, in the case rate of TB in the United States has declined during the
this case, the quality of evidence that treatment of TB disease and past several years, there remains a large reservoir of individuals
LTBI improve outcomes is high quality, so the overall quality of who are infected with Mtb. Without the application of improved
evidence rating was determined entirely by the accuracy study. diagnosis and effective treatment for LTBI, new cases of TB will
The decision to recommend for or against an intervention develop from within this group, which is therefore a major focus
was based upon consideration of the balance of desirable conse- for the control and elimination of tuberculosis [9].
quences (ie, benefits) and undesirable consequences (ie, harms, Mtb is transmitted from person to person via the airborne
burdens), quality of the evidence, patient values and preferences, route [10]. Several factors determine the probability of Mtb
cost, resource use, and feasibility. The subcommittees used open transmission: (1) infectiousness of the source patient—a posi-
discussion to arrive at a consensus for each of the recommen- tive sputum smear for acid-fast bacilli (AFB) or a cavity on chest
dations. An open voting procedure was reserved for situations radiograph being strongly associated with infectiousness; (2)
when the subcommittee could not reach consensus through host susceptibility of the contact; (3) duration of exposure of the
discussion, but this was not needed for any recommendation. contact to the source patient; (4) the environment in which the
The strength of a recommendation indicates the committee’s exposure takes place (a small, poorly ventilated space providing
certainty that the desirable consequences of the recommended the highest risk); and (5) infectiousness of the Mtb strain. In the
course of action outweigh the undesirable consequences. United States, among contacts of patients with TB disease eval-
A strong recommendation is one for which the subcommittee is uated during a contact investigation, about 1% have TB disease
certain, whereas a conditional recommendation is one for which themselves and 23% have a positive tuberculin skin test (TST)
the subcommittee is uncertain. Uncertainty may exist if the qual- without evidence of tuberculosis disease and are considered
ity of evidence is poor, there is a fine balance between desirable to have LTBI [11]. Those who are household contacts and are

e6 • CID 2017:64 (15 January)  •  Lewinsohn et al


exposed to patients who are smear positive have higher rates of protein precipitate of heat-inactivated tubercle bacilli (purified
both infection and disease [12]. Medical procedures that gener- protein derivative [PPD]–tuberculin). The TST has been the
ate aerosols of respiratory secretions, such as sputum induction standard method of diagnosing LTBI.
and bronchoscopy, entail significant risk for Mtb transmission The TST is administered by the intradermal injection of
unless proper precautions are taken [13]. 0.1  mL of PPD (5 TU) into the volar surface of the forearm
(Mantoux method) to produce a transient wheal. The test is
Initial Infection: Acquisition of Latent Mtb Infection interpreted at 48–72 hours by measuring the transverse diame-
After inhalation, the droplet nucleus is carried down the bron- ter of the palpable induration. TST interpretation is risk-strat-
chial tree and implants in a respiratory bronchiole or alveolus. ified [23]. A reaction of 5 mm or greater is considered positive
Whether or not inhaled tubercle bacilli establish an infection for close contacts of tuberculosis cases; immunosuppressed
depends on both host and microbial factors [14]. It is hypoth- persons, in particular persons with HIV infection; individu-
esized that, following infection, but before the development of als with clinical or radiographic evidence of current or prior
cellular immunity, tubercle bacilli spread via the lymphatics to TB; and persons receiving TNF blocking agents. A reaction of
the hilar lymph nodes and then through the bloodstream to ≥10 mm is considered positive for other persons at increased
more distant anatomic sites [15]. The majority of pulmonary risk of LTBI (eg, persons born in high TB incidence countries
tuberculosis infections are clinically and radiographically unap- and those with at risk of occupational exposure to TB) and
parent [16]. A  positive TST or IGRA result, most commonly, for persons with medical risk factors that increase the prob-
is the only indication that infection with Mtb has taken place. ability of progression from LTBI to TB (Figure 1). A reaction

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


Those who develop a positive TST are considered to have of 15  mm or greater is considered positive for all other per-
LTBI. It is estimated that, in the absence of treatment, approx- sons. Serious adverse reactions to PPD-tuberculin are rare.
imately 4%–6% of individuals who acquire LTBI will develop However, strong reactions with vesiculation and ulceration
active TB disease during their lifetime. The greatest risk of pro- may occur.
gression is during the first 2 years following exposure [11, 17]. The sensitivity of the TST, as measured in clinically well per-
The ability of the host to contain the organism is reduced in sons with previously treated tuberculosis, is high (95%–98%).
young (<4 years) children and by certain diseases such as silico- False-negative reactions occur more frequently in infants and
sis, diabetes mellitus, and diseases associated with immunosup- young children, early (<6–8 weeks) after infection, in persons
pression (eg, human immunodeficiency virus [HIV] infection), having recently received viral vaccination, in persons with
as well as by corticosteroids and other immunosuppressive clinical conditions associated with immunosuppression (eg,
drugs such as tumor necrosis factor alpha (TNF-α) inhibitors. HIV infection) or overwhelming illness (including extensive
In these circumstances, the likelihood of progression to TB dis- or disseminated tuberculosis), after recent viral and bacterial
ease is greater. For example, individuals who have a prior latent infections, and in association with treatment with immunosup-
infection with Mtb (not treated) and then acquire HIV infection pressive drugs (eg, high-dose corticosteroids, TNF inhibitors).
will develop TB disease at an approximate rate of 5%–10% per Test specificity of the TST is decreased among persons with
year (in the absence of effective HIV treatment) [18, 19]. prior BCG vaccination, especially those vaccinated postinfancy
and those with repeat vaccination. Similarly, persons living in
areas where nontuberculous mycobacteria are common are at
DIAGNOSTIC TESTS FOR LTBI
increased risk of having false-positive TST reactions. Repeated
The aim of testing for LTBI is to identify those who will benefit administration of TSTs cannot induce reactivity; however, a
from prophylactic therapy. At present, the likelihood of complet- repeat TST can restore reactivity in persons whose TST reac-
ing LTBI treatment is relatively modest. In some reports, only tivity has waned over time. Because of this “boosting phenom-
17%–37% of those eligible for LTBI therapy ultimately complete enon,” initial repeat testing is recommended for persons with a
the treatment course, with higher rates of completion associated negative TST who are to undergo periodic TST screening and
with shorter courses of therapy [20, 21]. Once therapy has been who have not been tested with tuberculin recently (eg, 1 year).”
initiated, completion rates are more favorable [22]. It is hoped This “2-step” testing, with a repeat TST within 1–3 weeks after
that better diagnostic tests, testing strategies, and treatment reg- an initial negative TST, is intended to avoid misclassification of
imens will allow for resources to be focused on patients who are subsequent positive TSTs as a TST conversion, indicating recent
most deserving of evaluation and treatment of LTBI and, there- infection, when they are actually a result of boosting.
fore, result in increased completion of therapy rates.
Benefits and Limitations of the TST
Tuberculin Skin Testing The benefits of the TST include its simplicity to perform (it
The tuberculin skin test (TST) detects cell-mediated immunity does not require a laboratory or equipment and can be done
to Mtb through a delayed-type hypersensitivity reaction using a by a trained healthcare worker in remote locations), its low

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e7


cost, no need for phlebotomy, the observation that it reflects a antigenic exposure. This response is characterized by the
polycellular immune response, and the foundation of well-con- release of cytokines, as well as further expansion of these cells.
trolled studies that support the use of the TST to detect LTBI Responses measured in current short-term IGRA assays reflect
and guide the use of prophylactic therapy [24]. In addition, the presence of these cells. Although it has been postulated
there are well-established definitions of TST conversion, which measurement of these short-term effectors might reflect recent
are particularly helpful when using the TST in the setting of infection and/or ongoing bacterial replication, current evidence
serial testing. does not support this hypothesis [41–43].
Limitations include the need for trained personnel to both
administer the intradermal injection and interpret the test, Commercially Available IGRAs
inter- and intrareader variability in interpretation, the need for Currently, there are 2 commercially available IGRA platforms that
a return visit to have the test read, false-positive results due to measure interferon-γ release in response to Mtb-specific antigens:
the cross-reactivity of the antigens within the PPD to both BCG the QuantiFERON TB Gold In Tube (QFT-GIT; Cellestis Limited,
and nontuberculous mycobacteria, false-negative results due to Carnegie, Victoria, Australia) and T-SPOT.TB test (T-SPOT,
infections and other factors, rare adverse effects, and compli- manufactured by Oxford Immunotec Ltd, Abingdon, United
cated interpretation due to boosting, conversions, and rever- Kingdom). The QFT-GIT measures IFN-γ plasma concentration
sions [24]. using an enzyme-linked immunosorbent assay (ELISA), while
the T-SPOT assay enumerates T cells releasing IFN-γ using an
Interferon-Gamma Release Assays enzyme-linked immunospot (ELISPOT) assay.

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


Overview of IGRAs
Until recently, the TST has been the only method to test for QuantiFERON Assays
latent infection with Mtb. Ideally, an improved diagnostic test The QFT-GIT method has been approved by the US Food
would specifically identify those with Mtb infection and would and Drug Administration (FDA) and has replaced the
delineate those at risk for disease progression. In this regard, the QuantiFERON-TB Gold (QFT-G) test. Whole blood (mini-
TST has well-known strengths and limitations [23, 25, 26]. The mum 3  mL) is drawn directly into heparinized tubes coated
IGRAs are newer tests to diagnose infection with Mtb. IGRAs with lyophilized antigen and agitated. In this case, peptides from
are in vitro, T cell–based assays that measure interferon gamma ESAT-6, CFP-10, and TB7.7 are found within the same tube.
(IFN-γ) release by sensitized T cells in response to highly spe- Two additional tubes are drawn as controls (mitogen control and
cific Mtb antigens. nil control). The mitogen control (phytohemagglutinin [PHA])
stimulates T-cell proliferation and ensures that viable cells are
Immune Basis of IGRAs present. After incubation for 16–24 hours at 37°C, plasma is col-
Like the TST, the IGRA is a reflection of the cellular immune lected from each tube and the concentration of IFN-γ is deter-
response. The discovery of antigens that have elicited robust mined for each by ELISA. The in-tube methodology requires
immune responses and are relatively specific for infection with no additional sample handling. Perhaps because of the nearly
Mtb has enabled the development of IGRA assays, which are immediate exposure of T cells to antigen, as well as the addition
more specific for Mtb infection than the TST [27], particularly of the TB7.7 peptide, the QFT-GIT may be more sensitive than
in the setting of BCG vaccination. Of particular interest has the QFT-G test. Studies reporting the sensitivity and specificity
been the RD-1 gene segment, a 9.5-kb DNA segment absent of the QFT-GIT test are provided in Supplementary Tables 1 and
from all strains of Mycobacterium bovis BCG but present in 2, respectively. The next generation of QFT (QFTPlus) has been
wild-type M.  bovis and Mtb [28]. This region, containing 11 introduced in Europe and is pending approval in the United
open reading frames, is responsible for the transcription and States. QFTPlus contains a tube of short peptides derived from
translation of a variety of antigenic proteins, including early CFP-10, which are designed to elicit an enhanced CD8 T-cell
secretory antigen (ESAT-6) [29–33] and culture filtrate protein response. There is no TB7.7 peptide. No published information
(CFP-10) [34–37]. Both antigens are absent from all attenuated is available to evaluate the performance of this test.
strains of M.  bovis (BCG strains) and most nontuberculous The QFT-GIT assay is considered positive if the difference
mycobacteria with the important exceptions of Mycobacterium between the IFN-γ concentration in response to the Mtb anti-
kansasii, Mycobacterium szulgai, Mycobacterium marinum [32, gens and the IFN-γ response to the nil control is ≥0.35 IU. In
38], and Mycobacterium leprae [39, 40]. addition, to control for high background in the nil control, the
IGRA assays are primarily a reflection of a CD4+ T-cell IFN-γ response to antigen must be 25% greater than the IFN-γ
immune response to these antigens. Immunologic memory concentration in the NIL control. An indeterminate response
is characterized by the clonal expansion of antigen-specific T defined as either a lack of response in the PHA control well
cells following exposure to an antigen. Effector memory T cells (IFN-γ concentration ≤0.5 IU) or a nil control that has a very
are defined by their capacity to respond rapidly to subsequent high background (IFN-γ concentration >8 IU).

e8 • CID 2017:64 (15 January)  •  Lewinsohn et al


T-SPOT.TB Assays malfunction, or technical errors might result in a poor mito-
The T-SPOT.TB assay is currently available in Europe, Canada, gen response. Here, it is reasonable to simply repeat the assay.
and has been approved for use in the United States with revised Second, a persistently diminished response to mitogen may
criteria for test interpretation. For the T-SPOT.TB assay, blood be a reflection of anergy. Thus, the reproducibility and details
(minimum 2  mL) is drawn into either a heparin or CPT Ficoll regarding the reason for an indeterminate result may provide
tube, and must be processed within 8 hours. More recently, this clinically useful information.
time has been extended to 32 hours if the “T-cell Xtend” additive is
used and the blood kept between 10°C and 25°C. Peripheral blood Reproducibility of IGRAs
mononuclear cells (PBMCs) are separated using density gradient Because IGRAs are predicated on in vitro release of cytokines
centrifugation, enumerated, and then added to microtiter wells at from stimulated cells, there is likely to be more variability in
2.5 × 105 viable PBMCs per well that have been coated with mon- these tests than those based on the measurement of a circu-
oclonal antibodies to IFN-γ (ELISPOT assay). Peptides derived lating substance such as sodium. There are at least 4 sources
from ESAT-6 and CFP-10 antigens are then added and the plate is of variability which are inherent in the IGRA: (1) the type of
developed following overnight (16–20 hours) incubation at 37°C. measurement itself (ie, ELISA or ELISPOT), (2) reproducibility
Cells are then washed away and “captured” IFN-γ is then detected of a complex biological reaction, (3) the natural variability of
via a sandwich capture technique by conjugation with secondary immune responses, and (4) variability introduced during the
antibodies hence revealing a “spot.” These spots are then enumer- course of test performance or manufacturing variances.
ated as “footprints” [44] of effector T cells [44, 45]. Reproducibility has been evaluated for both the QFT and

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


For the T-SPOT.TB assay, a positive response is based on T-SPOT assays. Although published information regarding
spot-forming units (SFU). Outside of the United States, if the currently available tests is limited [50, 51] the QFT-IT result
negative control well contains ≤5 SFU and there are >6 SFU was reported to have an 11% variance (http://www.access-
above the media nil control in either of the antigen wells, then data.fda.gov/cdrh_docs/pdf/P010033). Studies on within
this is considered positive. If the negative control well has ≥6 subject variability of the QFT-IT are limited and most were
SFU, then the antigen wells must be at least 2 times the negative performed in areas of the world where Mtb is endemic and
control well for a response to be considered positive. An invalid variability over time due to reinfection would be expected
response is defined as high background in the negative control [50, 51]. Recently, intrasubject variability of QFT-IT was
well (≥10 SFU) or if the positive control well is not responsive assessed using available plasma, and a discordance rate of 8%
to mitogen (PHA, <20 SFU). The FDA has published revised between the first and second tests was observed. While the
criteria for T-SPOT.TB interpretation in the United States, in variations were quantitatively modest, results at or near the
which a test is considered negative if there are ≤4 spots. Eight cutoff resulted in differing test results [52]. This variability
spots or greater is considered positive. Five, 6, and 7 spots are might spuriously change the test result (positive to negative
considered “borderline” and would be interpreted in conjunc- or negative to positive). Consequently, values at or near the
tion with the subject’s pretest probability of infection with Mtb. test cutoff should be interpreted with caution. Variability of
Studies reporting the sensitivity and specificity of the T-SPOT the T-SPOT was dependent on the strength of the response,
test are provided in Supplementary Tables 3 and 4, respectively. and varied from 4% in those with robust responses, to 22% in
those whose responses were close to the cutoff (http://www.
Indeterminate/Invalid IGRA Responses accessdata.fda.gov/scripts/cdrh/cfdocs/cftopic/pma/pma.
Unlike the TST, in which the results are interpreted categorically cfm?num=p070006).
based on the size of the reaction [46], the IGRAs currently have
a trichotomous outcome yielding a positive, negative, or inde- Boosting of IGRAs
terminate result (T-SPOT may also yield a borderline result as Initial studies found that repeat TST testing did not alter the
described above). As described above, an indeterminate/invalid IGRA response [53, 54]. However, more recent evidence [50,
IGRA can result from either a high background (nil) response or 51] suggests that the prior placement of a TST can boost an
from a poor response to positive control mitogen. Indeterminate IGRA, particularly in those individuals who were already IGRA
IGRA results are associated with immunosuppression [47–49], positive to begin with (ie, previously sensitized to Mtb or pos-
although they may occur in healthy individuals (studies report- sibly other mycobacteria). Additionally, it was found that this
ing the test characteristics of IGRAs in individuals with immu- could be observed in as little as 3 days post-TST administration,
nosuppression are provided in Supplementary Tables 5 and 6). and that the boosting effect may wane after several months [50,
With regard to those with a poor response to the positive con- 51]. While these data do not detract from the excellent overall
trol mitogen, there are at least 2 possibilities. First, the test may agreement that has been reported, they suggest when dual test-
not have been correctly performed. For example, errors in spec- ing is to be considered that the IGRA be collected either con-
imen collection, long delays in specimen processing, incubator currently or prior to TST placement.

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e9


Special Considerations morbidities [62] and has public health benefits, as each new
Because IGRAs rely on a functional assessment of viable lympho- case is likely to infect others. The consequences of failing to pre-
cytes, these tests require special attention to the technical aspects vent progression to active TB disease may be especially severe
of the test. This includes proper filling of the blood collection tube, in the young or immunocompromised host, in whom the dis-
proper mixing, timely transport to the laboratory, and timely pro- ease is more likely to be disseminated and elude discovery, and
cessing of the specimen. Additionally, for the laboratory, perfor- has a higher mortality rate. Failure to rapidly diagnose TB dis-
mance of cellular assays may pose unique challenges with regard to ease also poses a risk of widespread transmission in hospitals,
reagent storage and preparation as well as the separation of viable homeless shelters, and prisons.
cells. Finally, manufacturing problems such as endotoxin contam- Patients with LTBI have a 4%–6% lifetime risk of developing TB
ination can confound assays that depend on cellular activation. disease, with approximately half of these cases occurring follow-
ing recent exposure [11, 17]. Multiple placebo-controlled trials in
Benefits and Limitations of IGRAs adults and children with LTBI have shown that isoniazid reduces
The benefits of IGRAs include the use of antigens that are the subsequent development of TB disease in patients at high risk
largely specific for Mtb (ie, no cross-reactivity with BCG and of progression. As an example, in a trial of 28 000 individuals with
minimal cross-reactivity with nontuberculous mycobacteria), LTBI and radiographic evidence of healed tuberculosis, isonia-
the test can be performed in a single visit, and both the perfor- zid taken for 52 weeks reduced the subsequent development of
mance and reporting of results in a laboratory setting fall under TB disease from 14.3% to 3.6% [62]. Other groups in which the
the auspices of regulatory certification [24]. treatment of LTBI has been demonstrated to reduce the incidence

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


Limitations include cost, the need for phlebotomy (which of TB disease include household contacts of active TB patients
may be particularly challenging in children), complicated inter- [55, 63], native Alaskan communities [64], residents of mental
pretation due to frequent conversions and reversions and lack health facilities [65], persons with HIV infection [66–69], and
of consensus on thresholds, and inconsistent test reproducibility individuals treated with TNF inhibitors [70, 71]. These data can
[24]. The reproducibility of results is particularly problematic in be extrapolated to populations at low risk for progression (ie, no
the setting of serial testing. While some of this can be attributed risk factors) and intermediate risk for progression (ie, diabetes,
to results that fall near the cutoff, this is not always the case, and chronic renal disease, intravenous drug abuse); while the relative
current data does not provide specific guidance. Data on the benefit of treatment is probably similar in these lower risk popu-
effect of IGRA-guided therapy on prevention of TB disease is lations, the absolute benefit is almost certainly smaller due to the
limited, although one study demonstrated a roughly 84% reduc- lower baseline risk of progression to TB disease.
tion in TB disease among household contacts who received These studies provide high-quality evidence that treatment
IGRA-based preventive therapy [55]. Finally, several studies of LTBI reduces the incidence of TB disease in populations at
have reported an increased rate of indeterminate IGRA results in high risk for progression. However, they provide only moder-
children <5 years of age [47, 56–60] and one study described an ate-quality evidence that treatment of LTBI reduces the inci-
increased rate of indeterminate IGRA results among individuals dence of TB disease in populations at low or intermediate risk
with HIV infection and a CD4 count ≤200 cells/µL [61] for progression because the data are from high-risk populations.
Our recommendations for the diagnosis of LTBI reflect both
DIAGNOSTIC APPROACH: TESTING FOR the likelihood of infection (either likely or unlikely, based upon
SUSPECTED LTBI studies that used the TST to detect LTBI) and the risk of pro-
This section addresses how to test for LTBI. A complementary gression if infected (low; intermediate, RR 1.3–3; and high, RR
ATS/CDC/IDSA guideline that addresses who to screen for 3–10). This paradigm is summarized in Figure 1.
LTBI and how to treat LTBI is in development and forthcoming.
No definitive diagnostic test for LTBI exists. Our recommen- Question 1: Should an IGRA or a TST be performed in indi-
dations for diagnostic testing for LTBI are based upon the likeli- viduals 5  years or older who are likely to be infected with
hood of infection with Mtb and the likelihood of progression to Mtb, who have a low or intermediate risk of disease progres-
TB disease if infected, as illustrated in Figure 1. The recommen- sion, and in whom it has been decided that testing for LTBI
dations are summarized in Figure 2. As our literature searches is warranted?
failed to identify randomized trials or observational studies that
directly compared different diagnostic approaches and meas- Evidence
ured clinical outcomes, our recommendations are based upon In individuals who are likely to be infected with Mtb but at low
evidence about the accuracy of various tests combined with evi- or intermediate risk of disease progression, the sensitivity of
dence that treatment of LTBI improves clinical outcomes. IGRAs in the detection of Mtb infection has been consistently
There are 2 major benefits of treating LTBI: Treating LTBI reported as either equal (QFT; 81%–86%) or superior (T-SPOT;
prevents progression to active TB disease with its attendant 90%–95%) to the sensitivity of the TST (71%–82%) [47, 72–94]

e10 • CID 2017:64 (15 January)  •  Lewinsohn et al


when a final diagnosis of either microbiologically confirmed or are less likely with IGRAs than TST. This is important because
clinical TB is used as the reference standard. Individuals who false-positive results may lead to unnecessary treatment and its
are likely to be infected with Mtb include household contacts accompanying risks (ie, hepatotoxicity) [96–98]. To minimize
(studies reporting the test characteristics of IGRAs in contacts these risks, the guideline development panel chose to recom-
are provided in Supplementary Table  7), recent exposures of mend IGRA testing for individuals who received the BCG
an active case, mycobacteriology laboratory personnel, immi- vaccination.
grants from high-burden countries, and residents or employees In contrast, the accuracy of TST and IGRAs appears similar
of high-risk congregate settings. Individuals at low risk of pro- in those without a history of BCG vaccination. Despite the sim-
gression to TB include those with no risk factors, while those at ilar test characteristics, the guideline development committee
intermediate risk of progression to TB include those with diabe- chose to suggest IGRA testing over TST testing in such patients
tes, chronic renal failure, or intravenous drug abuse. because it was concerned about the reliability of a history of
In patients who are known to have received vaccination with having received or not received the BCG vaccination. Because
BCG, the specificity of IGRAs has also been consistently supe- many of the individuals who fall into the likely to be infected
rior to TST testing, presumably because IGRAs rely on responses with Mtb category are from regions of the world in which the
to antigens absent in BCG and many nontuberculous mycobac- BCG vaccination is routinely administered, the committee con-
teria. In contrast, among patients who have not received vac- cluded that individuals who are likely to be infected with Mtb
cination with BCG, the specificity of IGRAs and TST appears and provide history of not having received the BCG vaccination
similar. Meta-analyses estimate that the specificity of QFT–IT should be treated the same as those who provide a history of

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


to be >95%, whereas the specificity for TST is roughly 97% in having received the BCG vaccination, unless there is a reason
those with no prior exposure to BCG. The specificity is reduced to choose an alternate approach such as IGRA testing not being
to roughly 60% in those with a history of BCG vaccination [24]. available, being too costly, or being too burdensome.
Data for the commercially available T-SPOT are more limited. The recommendation to perform IGRA testing rather than
In German healthcare workers, specificity (using a cutoff of 6 TST testing is strong for those who have received the BCG vac-
spots) was reported at 97%, whereas in Korean adolescents the cination or who are not likely to return for TST read, reflecting
specificity was 85% [49]. In Navy recruits, specificity was 99% the guideline development committee’s certainty that avoiding
using the 8-spot cutoff [95]. the serious consequences of false-positive results and obtaining
Our confidence in the estimated test characteristics was a result to guide therapy outweigh the additional cost and need
moderate because many of the studies did not report whether to perform phlebotomy for IGRA testing. In contrast, the sug-
the subjects were consecutively enrolled. gestion to perform IGRA testing rather than TST testing on all
Recommendation 1a: We recommend performing an IGRA other patients who are likely to be infected with Mtb and have a
rather than a TST in individuals 5 years or older who meet the low or moderate risk of progressing to TB disease is conditional,
following criteria: (1) are likely to be infected with Mtb, (2) have reflecting the committee’s recognition that the choice should
a low or intermediate risk of disease progression, (3) it has been depend upon the clinical context as the test characteristics are
decided that testing for LTBI is warranted, and (4) either have similar. While the committee concluded that IGRA testing is
a history of BCG vaccination or are unlikely to return to have preferable in most patients, it recognized that TST testing may
their TST read (strong recommendation, moderate-quality evi- be more appropriate in a sizeable minority due to availability,
dence). Remarks: A TST is an acceptable alternative, especially feasibility, cost, or burden.
in situations where an IGRA is not available, too costly, or too
burdensome. Justification for Extending the Recommendation Down to
Recommendation 1b: We suggest performing an IGRA 5 Years of Age
rather than a TST in all other individuals 5 years or older who Young children are at increased risk of developing TB following
are likely to be infected with Mtb, who have a low or intermedi- infection and more likely to develop severe disease than older
ate risk of disease progression, and in whom it has been decided children and adults [99, 100]. This risk is highest in the young-
that testing for LTBI is warranted (conditional recommendation, est infants, diminishes with increasing age, and becomes equiv-
moderate-quality evidence). Remarks: A  TST is an acceptable alent with older children and adults at approximately 5  years
alternative, especially in situations where an IGRA is not availa- of age. Thus, children ≥5  years old have a similar risk of TB
ble, too costly, or too burdensome. as adults and display a similar disease spectrum. With respect
to Mtb infection, children aged ≥5  years possess a functional
Rationale immune response equivalent to that of adults. In addition, the
Accuracy studies indicate that IGRAs are more specific and results of existing studies of IGRA performance in children
equally or more sensitive than TST in individuals who have ≥5 years of age, albeit limited, are consistent with results of stud-
received the BCG vaccination; therefore, false-positive results ies of IGRA performance in adults. The sensitivity of IGRAs in

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e11


children with TB [56, 59, 101–105] and in older children who antigen is currently not available to drive an ongoing response
are household [106, 107] or school [108] contacts and the spec- that can be measured by IGRA), may reflect immune differences
ificity of IGRAs in children [102, 108] are comparable to those inherent in a delayed-type hypersensitivity versus blood assay,
of adults. For these reasons, it seems reasonable to extrapolate or may reflect exposures to nontuberculous mycobacteria. In
the results of studies of IGRA performance in healthy adults to low-risk populations, discordant tests are likely to be false posi-
children aged ≥5 years. tives [61, 111]. Clearly, more information is desirable regarding
which test best reflects productive infection and, therefore, best
Cautions and Limitations reflects the likelihood of disease progression.
While both IGRA and TST testing provide evidence for infec- The benefit of targeted testing for LTBI resides not in the
tion with Mtb, they cannot distinguish active from latent tuber- test employed, but in its programmatic use. We acknowledge
culosis. Therefore, the diagnosis of active TB must be excluded that programmatic considerations such as cost, test availability,
prior to embarking on treatment for LTBI. This is typically prevalence of BCG exposure in the target population, ability to
done by determining whether or not symptoms suggestive of reevaluate the patient 2–3  days after testing, and the training
TB disease are present, performing a chest radiograph and, if and expertise of program staff might all affect the decision to
radiographic signs of active tuberculosis (eg, airspace opacities, use IGRA- or TST-based evaluations.
pleural effusions, cavities, or changes on serial radiographs)
are seen, then sampling is performed and the patient managed Question 2: Should an IGRA or a TST be performed in indi-
accordingly. viduals 5  years or older who are likely to be infected with

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


Quantitative aspects of the tests are poorly understood. With Mtb, who have a high risk of progression to disease, and in
respect to the TST, the result is categorized as positive or nega- whom it has been decided that testing for LTBI is warranted?
tive and quantitative data are of limited utility, with the excep-
tion of recognition that a large (>15 mm) skin test reaction is Evidence
more likely to reflect infection with Mtb [109, 110]. The dichot- Individuals at high risk of progression to TB include those
omous characterization of the result, coupled with the fact that with HIV infection, an abnormal chest radiograph consist-
repeat testing is not recommended in the setting of a prior posi- ent with prior TB, or silicosis. It also includes those who are
tive test result, has resulted in a paucity of information about the receiving immunosuppressive therapy. Most data about the
variability of the TST result over time. With respect to IGRAs, accuracy of the TST and IGRA are from patients who are
measurement of IFN-γ over time may reflect inherent variabil- immunocompromised.
ity in the test result (the FDA accepts a variance of 11%) or true Studies have compared TST and IGRAs in the setting of
immunological variation due to alterations in the abundance of immunocompromise. Both diagnostic tests have diminished
Mtb antigens, exposure to other antigens, and/or the health and sensitivity in this setting. The sensitivity of IGRAs (QFT-IT and
nutritional status of the host. As an example, it is possible that T-SPOT) for detecting LTBI in individuals with HIV infection
a rise in IFN-γ might reflect ongoing exposure and/or growth has been estimated to be from 65% to 100% [112–114], while
of the bacteria. Alternatively, a rise in IFN- γ may reflect var- the sensitivity of TST is only 43% (25,85) when a final diagnosis
iability of the test. At present, there are insufficient data upon of either microbiologically confirmed or clinical TB is used as
which to base any recommendations for quantitative interpre- the reference standard. These limited data suggest that IGRAs
tation of IGRAs beyond those cut-points recommended by the are at least as sensitive as TST in the setting of HIV infection.
FDA. However, it is important to recognize that the optimal Studies have also compared IGRAs with TST in populations that
cut-points are controversial and results near the cut-point are were heterogeneous with respect to both the type of underlying
less reliable than results far above or below the cut-point. The immunocompromise and the reasons for testing. These stud-
results of IFN-γ testing should be reported quantitatively such ies demonstrated significant discord between TST and IGRA
that these immune correlates of the natural history of TB can results, but the source of the discordance has not been eluci-
be prospectively discerned and ultimately applied to clinical dated [61]. The panel’s confidence in the estimated test charac-
practice. teristics of IGRA and TST testing was moderate because it was
Discordance between TST and IGRA testing is common. not reported whether patients were consecutively enrolled or
Not surprisingly, TST-positive/IGRA-negative discordance is whether there was true diagnostic uncertainty.
often seen in persons with prior exposure to BCG. However, Recommendation 2: There are insufficient data to recom-
TST-positive/IGRA-negative discordant results where the TST mend a preference for either a TST or an IGRA as the first-line
is well over 15 mm have also been reported. The reasons for diagnostic test in individuals 5 years or older who are likely to
this delayed type hypersensitivity are not understood. It could be infected with Mtb, who have a high risk of progression to
relate to the possibility that discordance may reflect immune disease, and in whom it has been determined that diagnostic
responses that have occurred in the remote past (and where the testing for LTBI is warranted.

e12 • CID 2017:64 (15 January)  •  Lewinsohn et al


Rationale testing is more specific than TST testing and equally or more
The committee judged the body of evidence insufficient to sensitive than TST testing. We have no reason to suspect that
render a recommendation for either IGRA or TST testing in these relative test characteristics will be different among indi-
patients likely to be infected with Mtb who are at high risk for viduals who are unlikely to be infected with Mtb. However, it
progression to disease because the estimated test characteristics is likely that false-positive results are more common for both
were widely variable and derived from only a small subgroup of IGRAs and TST in populations with a lower prevalence of LTBI.
such patients (ie, immunocompromised patients). This is supported by a study of longitudinal testing of healthcare
As part of the discussion about which diagnostic test workers residing in areas of low TB prevalence, which found
to perform in patients likely to be infected who are at high that most conversions were false-positive results as evidenced
risk for progression to disease, many committee members by a negative result on repeat testing [116].
acknowledged that they perform a second test in their clinical The evidence provides low confidence in the estimated test
practices when such patients test negative; specifically, they characteristics in our population of interest because many of
perform a TST if an initial IGRA is negative or an IGRA if the estimates are based upon evidence from patients who are
an initial TST is negative. If the second test is positive, they likely to be infected with a high risk for progression rather
consider this evidence for infection with Mtb. Their practice is than patients who are unlikely to be infected, and many of the
not based upon empirical evidence, but rather, the following studies did not report whether subjects were consecutively
clinical rationale. A sensitive diagnostic test is important for enrolled.
individuals who are likely to be infected with Mtb and at high Guidelines recommend that persons at low risk for Mtb infec-

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


risk of progression, so that such individuals are less likely to tion and disease progression NOT be tested for Mtb infection.
receive false-negative results that will result in delayed diagno- We concur with this recommendation. However, we also rec-
sis and treatment. Performing a second diagnostic test when ognize that such testing may be obliged by law or credentialing
the initial test is negative is one strategy to increase sensitivity. bodies. If diagnostic testing for LTBI is performed in individu-
While this strategy to increase sensitivity may reduce the spec- als who are unlikely to be infected with Mtb despite guidelines
ificity of diagnostic testing, this may be an acceptable tradeoff to the contrary.
in situations in which it is determined that the consequences Recommendation 3a: We suggest performing an IGRA
of missing LTBI (ie, not treating individuals who may benefit instead of a TST in individuals 5 years or older (conditional
from therapy) exceed the consequences of inappropriate ther- recommendation, low-quality evidence). Remarks: A TST is an
apy (ie, hepatotoxicity). acceptable alternative in settings where an IGRA is unavailable,
too costly, or too burdensome.
Cautions and Limitations Recommendation 3b: We suggest a second diagnostic test
While both QFT and T-SPOT rely on the release of IFN-γ in if the initial test is positive in individuals 5 years or older
response to RD-1 antigens, limited data have suggested that (conditional recommendation, very low-quality evidence).
indeterminate results are more common for QFT-IT when Remarks: The confirmatory test may be either an IGRA or a
the CD4 count is <200 cells/µL than T-SPOT [115]. This may TST. When such testing is performed, the person is consid-
be the result of T-SPOT using a defined number of PBMCs, ered infected only if both tests are positive.
which may better, but not completely, normalize for the lack of
CD4+ T cells. By incorporating a measure of anergy into the Rationale
test (Mitogen control), IGRAs may more accurately allow the Current ATS/CDC and American Academy of Pediatrics guide-
clinician to discriminate a test that is negative from one that lines recommend that testing for LTBI not be performed in
is indeterminate (anergic by virtue of inadequate responses to individuals at low risk for infection with Mtb because the risk of
mitogen). isoniazid chemoprophylaxis may outweigh the potential benefit
[117]. Despite this, testing is often performed in conjunction
Question 3: Should an IGRA or a TST be performed in indi- with school enrollment, employee health testing, and other
viduals 5 years or older who are unlikely to be infected with institutional settings. In such patients, many conversions are
Mtb, but in whom it has been decided that testing for LTBI false results, which may lead to unnecessary therapy and, there-
is warranted? fore, unnecessary and age-related risk of hepatotoxicity.
The evidence indicates that false-positive results are frequent
Evidence (ie, more common than true-positive results) among individ-
There is a lack of direct evidence regarding the relative test uals who are unlikely to be infected with Mtb. Use of a more
characteristics of IGRA and TST testing in individuals who specific test may result in fewer false-positive results and, there-
are unlikely to be infected with Mtb. Indirect evidence from fore, fewer persons receiving unnecessary LTBI treatment and
individuals likely to be infected with Mtb indicates that IGRA being placed at risk for adverse outcomes. In addition to the risk

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e13


associated with isoniazid chemoprophylaxis, those with a posi- from a false-positive result [119]. As discussed above, there
tive test for LTBI often undergo additional screening, including are a number of sources of variability in the IGRA assay
a chest radiograph. Avoiding such unnecessary screening has related to laboratory technique such as sample agitation, time
both cost and health benefits. The desire for a more specific test elapsed prior to incubation, duration of incubation, agitation
favors IGRA testing over TST, according to evidence described technique, and blood volume that could result in variability
above from patients who are likely to be infected and who have around the cutoff value. In this instance, this variability may
a low or intermediate risk for progression. The notion of per- reflect the inherent variability of a biologic measurement, and
forming a second, confirmatory test following an initial posi- is the rationale behind the committee’s recommendation that
tive result is based upon the evidence that false-positive results quantitative values be reported. The optimal cut-points for
are common among individuals who are unlikely to be infected IGRA testing are controversial. While results close to the cut-
with Mtb and the committee’s presumption that performing a point tend to be less reliable than results substantially above
second test on those whose initial test was positive will improve or below the cut-point, this is not absolute; in many instances,
specificity. positive values well above the threshold were not repro-
The recommendations are both conditional because the duced in subsequent testing [116]. It is for this reason that
quality of evidence provided the committee with limited confi- the committee felt that quantitative guidance regarding the
dence in the estimated test characteristics of IGRAs and TST in interpretations of conversions and reversions in the context
individuals who are unlikely to be infected; therefore, the com- of healthcare worker screening could not be provided. Given
mittee could not be certain that the desirable consequences of the varied sources of IGRA variability [24], the committee

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


performing IGRAs instead of TST, or of performing a second thought that a positive test in a low-risk individual was likely
test following a positive result, outweigh the undesirable conse- to be a false-positive result, and recommended repeat testing.
quences in the vast majority of patients.
Question 4: Should an IGRA or a TST be performed in
Cautions and Limitations (Testing for TB in Healthcare Workers) healthy children <5 years of age in whom it has been decided
Traditionally, once an individual has had a positive TST, that testing for LTBI is warranted?
future use of the TST for screening is not recommended due Evidence
to the belief that the skin test will remain positive for life. The body of evidence regarding IGRA performance in young
In those who are TST negative, serial testing can be compli- children is limited. Compared with adults, a limited number of
cated by random variability, boosting (ie, increased reactions children have been enrolled in IGRA studies. Even fewer chil-
upon retesting due to immunological memory), conversions dren from nonendemic countries have been studied, and many
(ie, new reactions due to new infection), and reversions (ie, reports do not include a separate analysis of young children.
decreased reactions). Criteria for the placement and reading The sensitivity of IGRAs in young children with TB ranges
of the TST, as well as the effect of boosting with PPD, criteria from 52% to 100% when a final diagnosis of either microbio-
for TST conversion have been established. Receiver operating logically confirmed or clinical TB disease is used as the reference
characteristic (ROC) analysis has been used to establish crite- standard, which is comparable to adults [56, 59, 76, 102–105,
ria for positive and negative IGRA results in those thought to 120]. The sensitivity of the TST has been reported as equiva-
be unlikely to be infected or those with TB disease. However, lent or increased compared with IGRAs in children [56, 59, 76,
IGRAs have not proven to be the solution to the problem 101–105, 120], with young age associated with decreased IGRA
of false-positive results associated with serial testing in low positivity [107]. Important caveats to this comparison, however,
risk individuals. At present, there is insufficient information are that some studies used earlier, less sensitive versions of the
available to guide the establishment of definitive criteria for IGRA and results have been inconsistent. As examples of the
the conversion and possible reversion of IGRAs. The issue of inconsistencies, a study using an IGRA similar to a currently
interpreting IGRA conversions and reversions in the context available IGRA test demonstrated increased sensitivity of the
of serial testing has proven especially problematic. For exam- IGRA compared with TST in children aged <3 years, especially
ple, in a study of 216 Indian healthcare workers, a QFT con- among those coinfected with HIV and/or malnourished [121].
version rate of 12% and a reversion rate of 24% were observed, None of these studies were performed in nonendemic countries.
with many of these apparent changes occurring near the cutoff The specificity of IGRAs appears to be excellent in children
values [118]. A  longitudinal study involving 2563 in health- in the range of 90%–100% [122] according to a study conducted
care workers demonstrated an IGRA conversion (6%–8%) in in children who had nontuberculous mycobacteria. The study
those undergoing serial testing [116]. These rates were 6–9 found that IGRAs were more specific than TST in children with
times higher than that seen for the TST and were thought to nontuberculous mycobacterial disease [102].
have represent false conversions. Such studies have not yielded Our confidence in the estimated test characteristics of IGRAs
useful criteria that can be used to distinguish Mtb infection and TST in children is very low because most of the studies did

e14 • CID 2017:64 (15 January)  •  Lewinsohn et al


not report whether or not they enrolled consecutive patients, DIAGNOSTIC TESTS FOR TB
were not performed in nonendemic countries, and have pro- The diagnosis and management of TB disease rely on accurate
vided inconsistent results. laboratory tests, both for the benefit of individual patients and
Recommendation: We suggest performing a TST rather than the control of TB in the community through public health ser-
an IGRA in healthy children <5  years of age for whom it has vices. Therefore, laboratory services are an essential component
been decided that diagnostic testing for LTBI is warranted (con- of effective TB control at the local, state, national, and global
ditional recommendation, very low-quality evidence). Remarks: levels.
In situations in which an IGRA is deemed the preferred diag- In the United States, up to 80% of all initial TB-related labora-
nostic test, some experts are willing to use IGRAs in children tory work (eg, AFB smear and culture inoculation) is performed
over 3 years of age. in hospitals, clinics, and independent laboratories outside the
public health system, whereas >50% of species identification
Rationale and drug susceptibility testing (DST) is performed in public
The limited direct evidence described above suggests that the health laboratories [124]. Thus, effective TB control requires a
TST might be more sensitive than IGRAs in young children, network of public and private laboratories to optimize labora-
and IGRAs may be more specific than the TST, particularly in tory testing and the flow of information. Public health labora-
those given BCG. Because young children have a high risk for tory workers, as a component of the public health sector with a
progression to active TB disease, the committee believed that mandate for TB control, should take a leadership role in devel-
the sensitivity of the diagnostic test (ie, avoiding false-negative oping laboratory networks and in facilitating communication

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


results, missed opportunities to treat) is more important than among laboratory workers, clinicians, and TB controllers.
the specificity of the test (ie, avoiding false-positive results, Seven types of tests for the diagnosis of TB disease and detec-
unnecessary therapy). This is supported by the observations tion of drug resistance are performed within the tuberculosis
that the potential consequences of delayed treatment are high, laboratory system and recommended for optimal TB control
while the risk of hepatotoxicity is greatly reduced in young services (Table 2). These laboratory tests should be available to
children. An additional reason to favor TST testing over IGRA every clinician involved in TB diagnosis and management, and
testing in young children is that the management of the most to jurisdictional public health agencies charged with TB control.
at-risk young children (ie, young household contacts) depends For suspected cases of pulmonary TB, sputum smears for
upon the results of serial testing for infection, for which there AFB are correlated with the likelihood of transmission and then,
are no data for IGRAs in young children. for AFB smear–positive pulmonary cases, a nucleic acid ampli-
While there are theoretical benefits from IGRA testing (eg, fication assay provides rapid confirmation that the infecting
improved acceptance of LTBI therapy), these benefits have not mycobacteria are from the Mtb complex. Both sputum smears
been proven. Therefore, there is insufficient evidence that the for AFB and nucleic acid amplification tests (NAATs) should be
benefits of IGRA testing exceed the well-known limitations of available with rapid turnaround times from specimen collection.
the TST. For these reasons, it is too early to recommend replac-
ing the TST with IGRA testing. The recommendation is condi-
tional because the quality of evidence provided the committee Table 2.  Essential Laboratory Tests for the Detection of Mycobacterium
with limited confidence in the estimated test characteristics of tuberculosis
IGRAs and TSTs in children; therefore, the committee could
Test Time Required
not be certain that the desirable consequences of performing
I. Nucleic acid amplification test, 1d
IGRAs instead of TSTs outweigh the undesirable consequences
detection (NAAT-TB)
in the vast majority of patients. II. Nucleic acid amplification test, 1–2 d
resistance markers (NAAT-R)
Cautions and Limitations III. Acid-fast bacilli microscopy 1d
IV. Growth detection Up to 6–8 wk
In studies of young children that report rates of indeterminate  Liquid (average 10–14 d)
IGRA results, the frequency ranges from 0 to 35%, which is  Solid (average 3–4 wk)
generally higher than in studies that reported in adults. Several V. Identification of Mycobacterium 1 da
tuberculosis complex by DNA probe
studies have reported an increased rate of indeterminate IGRA or HPLC
results in children <5 years of age [47, 56–60]. As phlebotomy is VI. First-line drug susceptibility testing 1 to 2 wka
more difficult in young children, inability to perform the IGRA (liquid medium)
VII. Second-line and novel compound
due to insufficient blood volumes represents an additional prac-
drug susceptibility testing
tical limitation to IGRA testing in young children. A relatively   i. Liquid (broth-based) medium 1 to 2 wka
high incidence of failed phlebotomy has been documented in   ii. Solid (agar- or egg-based) medium 3 to 4 wka
some studies [106, 123]. Abbreviation: HPLC, high-performance liquid chromatography.aAfter detection of growth.

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e15


These tests facilitate decisions about initiating treatment for TB does not exclude pulmonary TB. Similarly, false-positive results
or a non-TB pulmonary infection, infection control measures are sufficiently common that a positive AFB smear result does
(eg, patient isolation), and, if TB is diagnosed, for reporting the not confirm pulmonary TB. Testing of 3 specimens is consid-
case and establishing priority for the contact investigation. ered the normative practice in the United States and is strongly
recommended by the CDC and the National Tuberculosis
DIAGNOSTIC APPROACH: TESTING FOR SUSPECTED Controllers Association to improve sensitivity given the per-
PULMONARY TB vasive issue of poor sample quality. Providers should request
Pulmonary TB is often first suspected on the basis of chest a sputum volume of at least 3  mL, but the optimal volume is
computed tomographic findings (Supplementary Table  8). 5–10 mL. Concentrated respiratory specimens and fluorescence
Randomized trials and controlled observational studies that microscopy are preferred.
directly compared diagnostic tests for pulmonary tuberculo-
sis and measured patient-important outcomes have not been Rationale
performed. Therefore, the recommendations in this section AFB smear microscopy can be performed in hours, is inex-
are based upon data that describe how accurate a diagnostic pensive, and is technically simple. Our committee predeter-
test is at confirming or excluding pulmonary TB, coupled with mined that AFB smear microscopy would be recommended if
the widely accepted knowledge that diagnosing pulmonary TB false-negative results occur <30% of the time (ie, sensitivity is
leads to therapy that dramatically improves patient-important ≥70%) and false-positive results occur <10% of the time (ie,
outcomes and reduces disease transmission [125, 126]. Finally, specificity is ≥90%). The likely outcome of a false-negative

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


it was the consensus of the committee that testing for LTBI result is additional diagnostic testing and/or delayed diagnosis
(TST or IGRA) cannot be used to exclude a diagnosis of TB due to the duration required for cultures to become positive,
and, hence, should not be used in the evaluation of those with whereas the likely outcome of a false-positive result is unnec-
suspected TB. essary therapy with its associated risk for hepatotoxicity. The
different thresholds for false results reflect the committee’s
Question 5: Should AFB smear microscopy be performed in recognition that the consequences of false-negative results are
persons suspected of having pulmonary TB? generally transient, whereas the consequences of false-pos-
Evidence itive results may be long lasting. In this case, the sensitivity
Performing 3 AFB smears confirms pulmonary TB with a and specificity of AFB smear microscopy were approximately
sensitivity of approximately 70% when culture-confirmed TB 70% and ≥90%, respectively, so AFB smear microscopy is
disease is the reference standard. The reason for performing recommended.
3 AFB smears is that each specimen increases sensitivity. The The recommendation is strong because the quality of evi-
sensitivity of the first specimen is 53.8%, which increases by dence provided the committee with moderate confidence in the
a mean of 11.1% by obtaining a second specimen. Obtaining estimated test characteristics of AFB smear microscopy, and the
a third specimen increases the sensitivity by a mean of only committee therefore felt certain that the desirable consequences
2%–5% (ie, false-positive results could exceed the additional of AFB smear microscopy (ie, an early presumptive diagnosis,
true-positive results obtained from a third specimen). initiation of therapy, and possibly less transmission) outweigh
The sensitivity of a first morning specimen is 12% greater the undesirable consequences (ie, cost, burden, effects of false
than a single spot specimen [127]. Concentrated specimens results) in the vast majority of patients.
have a mean increase in sensitivity of 18% compared with non-
concentrated specimens (using culture as the standard) and flu- Question 6: Should both liquid and solid mycobacterial cul-
orescence microscopy is on average 10% more sensitive than tures be performed in persons suspected of having pulmo-
conventional microscopy [128, 129]. The specificity of micros- nary TB?
copy is relatively high (≥90%), but the positive predictive value Evidence
(PPV) varies (70%–90%) depending upon the prevalence of A meta-analysis comparing 2 liquid culture methods with
tuberculosis versus nontuberculous mycobacterial disease [130, solid cultures found that both liquid culture methods were
131]. These accuracy studies provide moderate confidence in more sensitive (88% and 90%) than the solid culture method
the estimated test characteristics because many did not report (76%) when a combination of conventional solid media with a
having enrolled consecutive patients. broth-based method was the reference standard, and also had
Recommendation 5: We recommend that AFB smear micros- a shorter time to detection (13.2 and 15.2 days for liquid cul-
copy be performed, rather than no AFB smear microscopy, in all ture methods versus 25.8  days for the solid culture method)
patients suspected of having pulmonary TB (strong recommen- [132]. The specificity of all 3 methods exceeded 99%. Liquid
dation, moderate-quality evidence). Remarks: False-negative culture medium has a higher contamination rate than solid
results are sufficiently common that a negative AFB smear result culture medium due to the growth of bacteria other than

e16 • CID 2017:64 (15 January)  •  Lewinsohn et al


mycobacteria (4%–9% in the meta-analysis), which interferes When further stratified by whether the patient received treat-
with obtaining a valid culture result. This evidence provides ment, the specificity in untreated patients was 97%. The second
low confidence in the estimated test characteristics for 2 rea- meta-analysis reported an overall sensitivity of 85% and speci-
sons. First, there may be selection bias, as many of the stud- ficity of 97% and did not stratify according to the results of AFB
ies did not state whether they enrolled consecutive patients. smear microscopy [136]. There was significant heterogeneity
Second, there is indirectness, since the studies address the test in both meta-analyses. The third meta-analysis stratified the
characteristics of either test alone but the question is about the NAAT test characteristics in AFB smear microscopy–negative
tests combined. suspects according to clinical suspicion of tuberculosis [137].
Recommendation 6: We suggest that both liquid and solid It found that in AFB smear microscopy–negative individuals,
mycobacterial cultures be performed, rather than either cul- a positive NAAT result is beneficial when the clinical suspicion
ture method alone, for every specimen obtained from an of tuberculosis was intermediate or high (>30%) and a negative
individual with suspected TB disease (conditional recommen- NAAT result is of little use in excluding the presence of Mtb.
dation, low-quality evidence). Remarks: The conditional qual- This evidence provides low confidence in the estimated test
ifier applies to performance of both liquid and solid culture characteristics because there may be selection bias since many
methods on all specimens. At least liquid culture should be of the studies did not state whether they enrolled consecutive
done on all specimens as culture is the gold standard micro- patients with legitimate diagnostic uncertainty and there was
biologic test for the diagnosis of TB disease. The isolate significant inconsistency in the meta-analyses.
recovered should be identified according to the Clinical and Recommendation 7: We suggest performing a diagnostic

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


Laboratory Standards Institute guidelines and the American NAAT, rather than not performing a NAAT, on the initial res-
Society for Microbiology Manual of Clinical Microbiology piratory specimen from patients suspected of having pulmo-
[133, 134]. nary TB (conditional recommendation, low-quality evidence).
Remarks: In AFB smear–positive patients, a negative NAAT
Rationale makes TB disease unlikely. In AFB smear–negative patients
Mycobacterial culture is the laboratory gold standard for tuber- with an intermediate to high level of suspicion for disease,
culosis diagnosis, but the preferred type of cultures is uncer- a positive NAAT can be used as presumptive evidence of TB
tain. Liquid cultures alone are reasonably sensitive and highly disease, but a negative NAAT cannot be used to exclude pul-
specific, but limited by contamination. Solid cultures alone monary TB. Appropriate NAATs include the Hologic Amplified
are not sufficiently sensitive to reliably diagnose TB and gen- Mycobacteria Tuberculosis Direct (MTD) test (San Diego,
erally take longer to yield results; however, some Mtb isolates California) and the Cepheid Xpert MTB/Rif test (Sunnyvale,
are detected only on solid medium. Performing both liquid and California).
solid cultures likely improves the sensitivity of mycobacterial
cultures, while the liquid cultures provide a more rapid answer Rationale
and the solid cultures serve as a safeguard against contamina- Mycobacterial culture results require at least 1–2 weeks; there-
tion. The recommendation is conditional because the quality fore, rapid diagnostic tests that can be performed within hours
of evidence provided the committee with limited confidence in are desirable, such as AFB smear microscopy and diagnostic
the estimated test characteristics of the culture methods; there- NAAT. Diagnostic NAAT has the added advantage over AFB
fore, the committee could not be certain that the desirable con- smear microscopy of being able to distinguish Mtb from nontu-
sequences of performing both culture methods instead of only berculous mycobacteria. However, NAAT is appropriate only as
one method outweigh the undesirable consequences in the vast an adjunct to mycobacterial culture and AFB smear microscopy.
majority of patients. It is used as an adjunct to mycobacterial culture because it is not
sensitive enough to replace mycobacterial culture for diagnosis
Question 7: Should NAAT be performed on the initial res- and does not produce an isolate, which is needed for phenotypic
piratory specimen in persons suspected of having pulmonary DST. It is used as an adjunct to AFB smear microscopy because
TB? the test characteristics of NAAT are highly variable depending
Evidence upon the AFB smear results and clinical suspicion.
Three meta-analyses were identified and reviewed. The first In AFB smear–positive patients, NAAT yields false-negative
stratified the performance characteristics of NAAT based upon results only 4% of the time, indicating that it is reliable for exclud-
AFB smear results [135]. When AFB smear microscopy was ing pulmonary TB. In AFB smear–negative patients, clinical sus-
positive, the sensitivity and specificity of NAAT were 96% and picion needs to be considered. When there is an intermediate
85%, respectively. Most studies used culture as the reference to high level of suspicion for disease, NAAT yields sufficiently
standard. When AFB smear microscopy was negative, the sen- few false-positive results that a positive NAAT result can be used
sitivity decreased to 66% and the specificity increased to 98%. as presumptive evidence of TB and guide therapeutic decisions;

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e17


however, false-negative results are sufficiently common that accuracy study [146]. This assay, Xpert MTB/RIF, was >92%
NAAT cannot be used to exclude pulmonary TB. When the clin- sensitive and >99% specific for detection of rifampin resistance
ical suspicion for TB is low, NAAT is generally not performed when performed on a single specimen; the sensitivity increased
because false-positive results are unacceptably frequent. An to >97% when performed on 3 specimens [146]. Despite its
algorithm for interpretation and use of NAAT results in con- good sensitivity and specificity, the PPV of rapid molecular
junction with AFB smear results has been published [138]. DST for the detection of rifampin resistance is low in popula-
The recommendation is conditional because the quality of tions with a low prevalence of drug resistance (Supplementary
evidence provided the committee with limited confidence in Table 9) [147].
the estimated test characteristics of NAAT; therefore, the com- One of the assays also detects isoniazid resistance. It identi-
mittee could not be certain that the desirable consequences fied isoniazid resistance with a sensitivity and specificity of 84%
of performing NAAT (ie, promptly diagnosing TB disease and 99%, respectively, when culture-based DST is used as the
and initiating treatment), instead of not performing NAAT, reference standard. However, when the meta-analysis was per-
outweigh the undesirable consequences (ie, cost, false-posi- formed on a subgroup of studies that evaluated a newer version
tive results leading to unnecessary treatment, and false-neg- of the assay, the sensitivity increased to approximately 90%. This
ative results provided false reassurance) in the vast majority indicates that in appropriate subgroups of patients, false-posi-
of patients. tive and false-negative results occur in 1% and 10% of patients,
respectively. In contrast to rifampin resistance, the PPV of a
Cautions and Limitations test indicating isoniazid resistance is quite high, a reflection of

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


Laboratory-based diagnostic tests are not a replacement for isoniazid resistance being fairly prevalent in the United States
clinical judgment and experience. A  diagnosis of pulmonary (approximately 8%) [144].
tuberculosis can be made in the absence of laboratory confir- This evidence provides moderate confidence in the estimated
mation, especially in children [139]. Although there appears to sensitivities and specificities among patient subgroups with
be little increase in accuracy achieved by routinely performing increased rates of drug resistance. The confidence is moderate
NAAT on multiple specimens rather than on a single speci- instead of high because the absence of reporting that patients
men, some clinicians may find it beneficial in the diagnosis were enrolled consecutively suggests that there is a risk of bias.
of individual patients [140, 141]. As an example, the presence Recommendation 8: We recommend performing rapid
of inhibitors can cause false-negative results for some NAATs molecular DST for rifampin with or without isoniazid using
[142] and, therefore, if a specimen has a positive AFB smear the respiratory specimens of persons who are either AFB smear
result and a negative NAAT result, evaluation of the sample for positive or Hologic Amplified MTD positive and who meet one
the presence of inhibitors should be considered if the NAAT of the following criteria: (1) have been treated for tuberculosis
being used is subject to inhibition. If inhibitors are detected, in the past, (2) were born in or have lived for at least 1  year
collection of a new specimen for NAAT should be considered. in a foreign country with at least a moderate tuberculosis inci-
The recommendation for use of NAATs is based on stud- dence (≥20 per 100 000) or a high primary MDR-TB prevalence
ies of commercial test kits. The data on in-house tests show (≥2%), (3) are contacts of patients with MDR-TB, or (4) are
even greater heterogeneity [143]. If in-house tests are to be HIV infected (strong recommendation, moderate-quality evi-
used, they should be validated and be shown to have analyti- dence). Remarks: This recommendation specifically addresses
cal performance accuracy comparable to or better than that of patients who are Hologic Amplified MTD positive because the
commercial tests. Hologic Amplified MTD NAAT only detects TB and not drug
resistance; it is not applicable to patients who are positive for
Question 8: Should rapid molecular drug susceptibility test- types of NAAT that detect drug resistance, including many line
ing for isoniazid and rifampin be performed as part of the probe assays and Cepheid Xpert MTB/RIF.
initial diagnostic evaluation for all patients suspected of hav-
ing pulmonary TB or only in selected subgroups? Rationale
Evidence Conventional, culture-based DST is the laboratory gold stand-
Rapid molecular DST can be performed via line probe or ard [134, 148, 149]. It is performed routinely any time Mtb com-
molecular beacon assays. We evaluated systematic reviews plex is isolated in culture. Drug susceptibility testing is essential
with meta-analyses of 2 line probe assays [144, 145]. Both line because treatment success for patients with MDR-TB (can
probe assays detected rifampin resistance with a sensitivity and reach 75% or higher [150, 151]) is dependent upon patients
specificity of ≥97% and ≥98%, respectively, when conventional, being treated with an effective antimicrobial regimen [152].
culture-based DST was used as the reference standard. More An important limitation of culture-based DST, however, is that
recently, a molecular-beacon based method for rapid rifampin it can take >2 weeks to grow the isolate that is necessary for
resistance detection was evaluated in a large international testing.

e18 • CID 2017:64 (15 January)  •  Lewinsohn et al


Rapid molecular DST addresses this limitation. It can be per- Other assays for rapid detection of drug resistance using
formed within hours, enabling earlier initiation of an appropri- alternative molecular techniques (eg, automated real-time pol-
ate antimicrobial regimen. Rapid molecular DST is an adjunct ymerase chain reaction [PCR] with sequencing, loop-mediated
and not a replacement for culture-based DST because it only isothermal amplification [LAMP]) are being developed. These
evaluates susceptibility to rifampin and occasionally isonia- assays are promising, but are not yet commercially available
zid. Nonetheless, detection of rifampin resistance is helpful to [146, 159, 160]. The data on in-house tests show substantial het-
clinicians because it is a good surrogate for MDR-TB in loca- erogeneity [161]. If in-house tests are to be used, they should
tions where rifampin monoresistance is uncommon. However, be validated and shown to have performance accuracy at least
an important limitation is that the PPV is expected to be lower comparable to that of commercial tests. The same cautions also
in the United States than in areas where rifampin resistance is apply to new commercial assays that may become available in
more common [153–155]. the near future.
The committee recommends rapid molecular DST only for Some clinicians and health departments may opt for broader
subgroups in which drug resistance is more likely, as the PPV use of the molecular detection of drug resistance assays than
for rifampin resistance testing is low in populations with a low recommended above, especially in regions where MDR-TB is
prevalence of drug resistance. Examples of appropriate persons more common. Because the prevalence of rifampin resistance
for testing include those who are NAAT or AFB smear positive (and therefore MDR-TB) is low in the United States, the PPV
and meet one of the following criteria: (1) have been treated for of Xpert MTB/RIF and other assays for rifampin resistance will
tuberculosis in the past, (2) were born in or have lived for at be lower than in settings where Xpert MTB/RIF has been pre-

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


least 1 year in a foreign country with at least a moderate tuber- dominantly studied. Therefore, confirmation of a positive test
culosis incidence (≥20 per 100 000) or a high primary MDR-TB result for rifampin resistance has been recommended [147]. To
prevalence (≥ 2%), (3) are contacts of patients with MDR-TB, or confirm a positive result, genetic loci associated with rifampin
(4) are HIV infected [154, 156–158]. resistance (to include rpoB), as well as isoniazid resistance (to
The sensitivity and specificity of rapid molecular DST for include inhA and katG), should be sequenced to assess for
detecting rifampin resistance are both >97%, indicating that MDR-TB. If mutations associated with rifampin resistance are
false-positive and false-negative results occur <3% of the time; confirmed, rapid molecular testing for other known mutations
thus, rapid molecular DST can be used confirm or exclude associated with drug resistance (to first-line and second-line
rifampin resistance in respiratory specimens. The sensitivity and drugs) is needed for healthcare providers to select an opti-
specificity of rapid molecular DST for detecting isoniazid resist- mally effective treatment regimen. All molecular testing should
ance are estimated to be 90% and 99%, respectively, indicating prompt growth-based DST.
that false-positive and false-negative results occur roughly 1% Alternative methods for rapid molecular DST are being
and 10% of the time, respectively; thus, rapid molecular DST developed and other technologies are likely to become available
can be used to confirm isoniazid resistance in respiratory spec- in the near future (eg, automated real-time PCR with sequenc-
imens, but not exclude it. ing, LAMP) [162]. It is possible that these techniques will be
The recommendation is strong because the moderate-quality sufficiently sensitive to be used for AFB smear–negative spec-
evidence provided the committee with sufficient confidence in imens. Laboratories in the United States should only use tests
the test characteristics to be certain that the benefits of rapid approved by the FDA or tests that have been produced and val-
molecular DST (ie, early identification of possible MDR-TB and idated in accord with applicable FDA and Clinical Laboratory
initiation of an appropriate antimicrobial regimen) outweigh Improvement Amendments regulations.
the costs and burden of testing in the overwhelming majority of
patients who have increased risk for drug resistance. Question 9: Should respiratory specimens be collected from
children with suspected pulmonary TB disease?
Cautions and Limitations Evidence
Line probe and molecular beacon assays have not been suffi- Respiratory specimens that can be collected from children
ciently validated for use on specimens other than respiratory include gastric aspirates; sputum collected by spontaneous
specimens. The recommendation for line probe assays and expectoration, induction, or nasopharyngeal aspiration; and
molecular beacon on respiratory specimens is based upon bronchoalveolar lavage (BAL). Gastric aspirates involve intu-
studies of commercial test kits, only one of which is currently bating the stomach after an overnight fast to collect swallowed
approved by the FDA: the molecular beacon–based method, sputum before the stomach empties. Collection of specimens
Xpert MTB/RIF. It is the only FDA-approved assay and it inte- on 3 consecutive mornings from patients with suspected pul-
grates diagnosis of TB and detection of rifampin resistance. If monary TB provides a diagnostic yield of up to 40%–50%, with
this test is used for the diagnosis of TB, a rifampin resistance higher yields for infants (up to 90%), symptomatic children,
result is automatically provided regardless of patient risk. and children with extensive disease (up to 77%), using a clinical

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e19


diagnosis of TB disease in a low prevalence country as criteria molecular fingerprinting between the isolates collected from
for the diagnosis of TB disease [163–165]. Meticulous attention children with culture-proven TB compared to their presumed
to detail during the collection and processing of the specimen source case [169, 170]. In contrast, no discordance was found
can improve yield (details are provided at http://www.currytb- between pediatric TB cases and their presumed source cases
center.ucsf.edu/pediatric_tb/). Sputum collected from children from 2000 to 2004 in Houston [171].
by nasopharyngeal aspiration or sputum induction with a bron- The recommendation is conditional because the moderate
chodilator has a yield of 20%–30% [166], whereas BAL in chil- quality of evidence provided the committee with insufficient
dren with pulmonary TB has a yield of 10%–22% [167]. These confidence in the estimated diagnostic yield; thus, the com-
estimates of diagnostic yield are based upon moderate-quality mittee felt uncertain that a diagnosis was rendered frequently
evidence—accuracy studies for which it was not documented enough that the desirable consequences of collecting respira-
whether the subjects were enrolled consecutively. tory specimens (ie, confirming the diagnosis of TB, obtaining
Recommendation 9: We suggest mycobacterial culture of an isolate for DST) outweigh the undesirable consequences (ie,
respiratory specimens for all children suspected of having pul- cost, burden, effects of false results) in the vast majority of chil-
monary TB (conditional recommendation, moderate-quality dren with suspected pulmonary TB.
evidence). The highest yields for gastric aspirates are in the youngest
Remarks: In a low incidence setting like the United States, it infants, in children with extensive or symptomatic disease, and
is unlikely that a child identified during a recent contract inves- for the first gastric aspirate collected. While there are situations
tigation of a close adult/adolescent contact with contagious where a presumed source case is not the child’s true source case,

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


TB was, in fact, infected by a different individual with a strain in the case of a very recent contact investigation of a house-
with a different susceptibility pattern. Therefore, under some hold-type contact with pan-susceptible disease, performing
circumstances, microbiological confirmation may not be nec- only one gastric aspirate or relying on the source case suscep-
essary for children with uncomplicated pulmonary TB identi- tibility may be appropriate. For infants, immunocompromised
fied through a recent contact investigation if the source case has hosts, children with extensive, disseminated, or extrapulmo-
drug-susceptible TB. nary disease, exposure to other potential source cases, or risk
of drug-resistance, respiratory specimens should be collected.
Rationale Studies comparing the yield of gastric aspirates to sputum have
Despite the observation that less than half of pediatric spec- shown discrepent results. Selection of an appropriate respira-
imens yield a positive culture, the committee judged that the tory specimen (i.e., gastric aspirates, spontaneous or induced
desirable consequences of mycobacterial cultures of respiratory sputa, or rarely bronchoalveolar lavage) should be based upon
specimens outweigh the undesirable consequences of specimen the expertise of the clinic and provider, the patient’s age and
collection in children for several reasons. First, a positive myco- developmental level, and the likelihood of an alternative diagno-
bacterial culture is likely to be reassuring to parents and staff sis. Most investigators have not found increased yield for bron-
that the diagnosis of tuberculosis is correct. Second, cultures are choalveolar lavage compared to gastric aspirates. Bronchoscopy
necessary for DST, which is particularly important in situations should be reserved for situations where an alternative diagnosis
in which TB drug resistance is prevalent. Third, suspectibility is being considered or when the anatomy is unclear.
data are not always available from the presumed source case.
Finally, after-the-fact culture collection in the face of treatment Cautions and Limitations
failure may have even lower yield than sampling a drug-na- Gastric aspirates are rarely AFB smear positive and the yield
ive child. Specimens that can be used for mycobacterial culture of cultures is suboptimal in children with pulmonary TB; thus,
include gastric aspirates, sputum, and BAL; the panel decided gastric aspirate culture results are helpful only if they are pos-
that there was insufficient evidence to advocate one collection itive. Negative results should not dissuade the provider from
method over another. empirically treating tuberculosis in children in the appropriate
With respect to the need for DST, overtreatment for pre- clinical setting. Gastric aspirate, sputum induction, and naso-
sumed drug-resistant TB may lead to unnecessary toxicities and pharyngeal aspiration in children are not comfortable and not
cost, while undertreatment due to unidentified drug resistance without financial cost. The procedures have modest risk (bleed-
may lead to treatment failure, risk of dissemination, and even ing from the nose, bronchospasm, airway intubation).
death. While it is tempting to avoid culture collection from the
child contact when a putative source case is identified (espe- Question 10: Should sputum induction or flexible broncho-
cially when susceptibility results are already available), prior scopic sampling be the initial respiratory sampling method
case series indicate that 2%–10% of children have susceptibility for adults with suspected pulmonary TB who are either una-
patterns that differ from the presumed source case [168] and ble to expectorate sputum or whose expectorated sputum is
more recent US studies have found up to 15% discordance of AFB smear microscopy negative?

e20 • CID 2017:64 (15 January)  •  Lewinsohn et al


Evidence committee to be absolutely certain that the balance of desira-
We identified 6 studies [172–177] that compared the diagnostic ble to undesirable consequences favors induced sputum over
yield of induced sputum with the yield of specimens obtained bronchoscopy.
by flexible bronchoscopy, using a positive mycobacterial culture
or evidence of a response the therapy as criteria for the diag- Question 11: Should flexible bronchoscopic sampling be per-
nosis of pulmonary TB. Five of the 6 studies demonstrated a formed in adults with suspected pulmonary TB from whom a
higher yield from induced sputum than bronchoscopy, with respiratory sample cannot be obtained via induced sputum?
the remaining study [176] demonstrating a similar yield. The Evidence
diagnostic yield of induced sputum increases with multiple Numerous studies reported the diagnostic yield of respiratory
specimens, with detection rates by AFB smear microscopy of specimens obtained by flexible bronchoscopy, using a positive
91%–98% and mycobacterial culture of 99%–100% reported mycobacterial culture or evidence of a response the therapy as
when 3 or more specimens are obtained [178]. criteria for the diagnosis of pulmonary TB [172–175, 178–182].
Two cost-analysis studies favored sputum induction over Generally speaking, bronchoscopic sampling appears to have a
bronchoscopy [172, 174]. In the first study, direct costs for diagnostic yield of 50%–100% when based on culture in patients
bronchoscopy measured in Canadian dollars were $187.60, suspected of having pulmonary TB. This yield appears unaffected
compared with $22.22 for sputum induction [172]. In the sec- by HIV infection, with bronchoscopy leading to an early pre-
ond study, induced sputum was about one-third the cost of flex- sumptive diagnosis of TB in 34%–48% of HIV-infected patients
ible bronchoscopy, and the most cost-effective strategy was 3 according to 2 studies [183, 184]. In one study, bronchial washings

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


induced sputa without bronchoscopy [174]. had the same culture yield (95%) as BAL but higher frequency of
Our confidence in the accuracy of the study results is low positive AFB smears (26% vs. 4%) [235]. Bronchoscopic brush-
because there was a risk of bias and indirectness. With respect ings yield AFB smear–positive results in 9%–56% [180, 185].
to risk of bias, most of the studies did not report whether or not Transbronchial biopsy (TBB) provides histopathologic find-
consecutive patients were enrolled. Supporting this concern, ings suggestive of pulmonary TB in 42%–63% of specimens
the variability of prevalence among studies suggests that the from smear-negative HIV-uninfected patients [183, 186]. HIV-
degree of diagnostic uncertainty likely differed among studies. infected patients are less likely (9%–19%) to demonstrate gran-
With respect to indirectness, there appeared to be indirectness ulomas on TBB [183, 186], although in 2 studies TBB was the
of the intervention because the studies varied in the number of exclusive means of diagnosing pulmonary TB in 10%–23% of
specimens collected (from 1 to 3 per patient), the concentra- patients [183, 184].
tions of hypertonic saline, the type of nebulizers, and the cul- Our confidence in the accuracy of these estimated diagnostic
ture techniques. yields is very low because most of the studies did not report
Recommendation 10: We suggest sputum induction rather whether or not consecutive patients were enrolled, the range of
than flexible bronchoscopic sampling as the initial respiratory reported diagnostic yields is wide, and the studies varied in how
sampling method for adults with suspected pulmonary TB who specimens were collected (bronchial aspirates and/or BAL and/
are either unable to expectorate sputum or whose expectorated or bronchial brushings and/or TBB) and the culture techniques.
sputum is AFB smear microscopy negative (conditional recom- Recommendation 11: We suggest flexible bronchoscopic
mendation, low-quality evidence). sampling, rather than no bronchoscopic sampling, in adults
with suspected pulmonary TB from whom a respiratory sample
Rationale cannot be obtained via induced sputum (conditional recommen-
Induced sputum has equal or greater diagnostic yield than dation, very low-quality evidence). Remarks: In the committee
bronchoscopic sampling, has fewer risks, and is less expensive. members’ clinical practices, BAL plus brushings alone are per-
These features all favor induced sputum as the initial respira- formed for most patients; however, for patients in whom a rapid
tory sampling method in patients with suspected pulmonary diagnosis is essential, transbronchial biopsy is also performed.
TB who are either unable to expectorate sputum or whose
expectorated sputum is AFB smear microscopy negative. The Rationale
committee recognizes that a potential advantage of bronchos- The committee judged that the desirable consequences of bron-
copy over sputum induction is the possibility of making a rapid choscopic sampling outweigh the undesirable consequences
presumptive diagnosis of tuberculosis by performing biopsies among patients with suspected pulmonary TB from whom
and identifying typical histopathologic findings, but felt that respiratory samples could not be obtained noninvasively. The
the balance of the upsides to downsides of induced sputum most important reason to perform bronchoscopy in a patient
outweighed that of bronchoscopic sampling. The recommen- with possible pulmonary TB is to differentiate TB disease from
dation is conditional because the quality of evidence does alternative diseases. Another reason to perform bronchoscopy
not provide sufficient confidence in the study results for the is to obtain specimens for cultures that provide isolates for DST.

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e21


Empiric treatment for presumed drug-resistant TB may lead Rationale
to unnecessary toxicities and cost if the patient actually has The rationale for postbronchoscopy sputum collection is the
drug-sensitive TB, while empiric treatment for drug-sensitive same as that described above for the bronchoscopic collection
TB may lead to treatment failure, risk of dissemination, and even of respiratory specimens.
death if the patient actually has drug-resistant TB. Moreover,
delayed diagnosis of drug resistance will prolong therapy and Question 13: Should flexible bronchoscopic sampling be per-
increase risk of default. Bronchoscopy also provides the oppor- formed in adults with suspected miliary TB and no alterna-
tunity of obtaining a rapid presumptive diagnosis by identifying tive lesions that are accessible for sampling whose induced
histopathologic findings consistent with tuberculosis. These sputum is AFB smear microscopy negative or from whom a
benefits of bronchoscopic sampling were thought to outweigh respiratory sample cannot be obtained via induced sputum?
the risks of bronchoscopy and the accompanying sedation, as
well as the costs and burdens. Evidence
The recommendation is conditional, reflecting the guide- Specimens obtained via bronchoscopy can undergo AFB
line development committee’s uncertainty that the desirable smear microscopy, mycobacterial culture, NAAT, and his-
consequences of bronchoscopy outweigh the undesirable topathological analysis. There is a paucity of evidence
consequences in many situations. Reasons for the commit- regarding the diagnostic characteristics of various types of
tee’s uncertainty included the highly variable estimates of bronchoscopic specimens (ie, washings, BAL, brushings,
diagnostic yield, the very low quality of evidence, recogni- TBB) obtained from patients with possible miliary TB. It has

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


tion that bronchoscopy is an invasive procedure and the risk been reported that bronchial washings, brushings, and TBB
of harm varies according to the patient’s clinical condition, have diagnostic yields of 14% [234], 27%–78% [181, 189,
and recognition that the feasibility of timely bronchoscopy 190], and 32%–75% [181, 189, 190], respectively. The diag-
varies according to the clinical setting. For example, in the nostic yield of BAL has not been reported. Our confidence
context of a public health clinic, the benefits of obtaining a in the accuracy of the estimated diagnostic yields is very low
bronchoscopy may not justify the thousands of dollars that because most of the studies did not report whether or not
it will cost due to professional fees, hospital charges, pathol- consecutive patients were enrolled (the prevalence of miliary
ogy costs, and laboratory fees, or the days to weeks of delays TB ranged from 55% to 90% in the studies, suggesting that the
that will be necessary to refer the patient to a pulmonologist degree of diagnostic uncertainty differed among the studies),
for bronchoscopy. In some situations, the potential harm the ranges of diagnostic yields were wide for both brushings
associated with delayed diagnosis may warrant empiric ini- and TBB, reflecting the variable results reported by the indi-
tiation of therapy based upon a reasonable suspicion of TB vidual studies, and the studies varied in the technique used to
disease. perform the sampling (particularly TBB) and the number of
specimens collected.
Question 12: Should postbronchoscopy sputum specimens Recommendation 13: We suggest flexible bronchoscopic
be collected from adults with suspected pulmonary TB? sampling, rather than no bronchoscopic sampling, in adults
Evidence with suspected miliary TB and no alternative lesions that
Postbronchoscopy sputum specimens are typically sent for AFB are accessible for sampling whose induced sputum is AFB
smear microscopy and mycobacterial culture. Postbronchoscopy smear microscopy negative or from whom a respiratory
AFB smears have a diagnostic yield of 9%–73% and postbro- sample cannot be obtained via induced sputum (condi-
nchoscopy mycobacterial cultures have a yield of 35%–71% tional recommendation, very low-quality evidence). Remarks:
according to multiple studies [182, 185, 187, 188]. In HIV- Bronchoscopic sampling in patients with suspected miliary
infected patients, the yield of postbronchoscopy sputum cul- TB should include bronchial brushings and/or TBB, as the
tures was 80% in a single study [186]. Our confidence in the yield from washings is substantially less and the yield from
accuracy of the estimated diagnostic yields is low because most BAL unknown. For patients in whom it is important to pro-
of the studies did not report whether or not consecutive patients vide a rapid presumptive diagnosis of tuberculosis (ie, those
were enrolled and the diagnostic yields reported varied greatly who are too sick to wait for culture results), TBB is both nec-
as indicated by the wide ranges described above. essary and appropriate.
Recommendation 12: We suggest that postbronchoscopy
sputum specimens be collected from all adults with suspected Rationale
pulmonary TB who undergo bronchoscopy (conditional recom- The rationale for bronchoscopic sampling in individuals with sus-
mendation, low-quality evidence). Remarks: Postbronchoscopy pected miliary TB and no alternative lesions that are accessible for
sputum specimens are used to perform AFB smear microscopy sampling (eg, enlarged lymph nodes or draining lesions) whose
and mycobacterial cultures. induced sputum is AFB smear negative or from whom a respiratory

e22 • CID 2017:64 (15 January)  •  Lewinsohn et al


sample cannot be obtained via induced sputum is essentially the 2 case series that included patients with tuberculous menin-
same as that described above for patients with suspected pulmo- gitis found that mortality due to disseminated tuberculosis is
nary TB from whom a respiratory sample cannot be obtained via 20% [193, 194]. Nevertheless, mortality is clearly improved
induced sputum. That is, it is important to differentiate miliary TB with a large magnitude of effect. This evidence constitutes
from other diseases and also to obtain specimens for mycobacte- moderate-quality evidence that treatment of extrapulmonary
rial culture because cultures provide isolates for DST, which may TB improves mortality because there are observational stud-
prevent unnecessary drug toxicities and cost, mitigate treatment ies with a very large magnitude of effect, but the increase in
failure, and reduce the risk of dissemination and death. Moreover, confidence in the results due to the large magnitude of effect is
bronchoscopy also provides the opportunity of obtaining a rapid mitigated by the indirectness of the population.
presumptive diagnosis by identifying histopathologic findings con-
sistent with tuberculosis. These benefits outweigh the risks of both Question 14: Should cell counts and chemistries be per-
bronchoscopy and the accompanying sedation. formed on amenable (ie, liquid) specimens collected from
The recommendation is conditional, reflecting the guide- sites of suspected extrapulmonary TB?
line development committee’s uncertainty that the desirable Evidence
consequences of bronchoscopy outweigh the undesirable con- We identified no studies that reported the sensitivity and spec-
sequences and its recognition that the balance of desirable ificity of cell counts and chemistries in the identification of
and undesirable consequences depends upon clinical context. extrapulmonary TB. Therefore, the committee used its collec-
Reasons for the committee’s uncertainty included the variable tive clinical experience to inform its recommendation. Clinical

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


estimates of the diagnostic yield and very low quality of evi- experience constitutes very low-quality evidence.
dence. Clinical considerations that may affect the balance of Recommendation 14: We suggest that cell counts and chem-
desirable and undesirable effects include the patient’s condi- istries be performed on amenable fluid specimens collected
tion (ie, bronchoscopy is an invasive procedure and the risk of from sites of suspected extrapulmonary TB (conditional rec-
harm varies according to the patient’s clinical condition) and ommendation, very low-quality evidence). Remarks: Specimens
the availability, feasibility, and cost of timely bronchoscopy in a that are amenable to cell counts and chemistries include pleural,
particular clinical setting. cerebrospinal, ascitic, and joint fluids.

DIAGNOSTIC APPROACH: TESTING FOR SUSPECTED Rationale


EXTRAPULMONARY TB Cell counts and chemistries can be performed in hours, are
Randomized trials and controlled observational studies that inexpensive, and are technically simple. Any risks are related to
directly compared diagnostic tests or approaches for extrapul- the sampling procedure. Although their sensitivity and speci-
monary TB and measured patient-important outcomes have not ficity for extrapulmonary TB have not been reported, the com-
been performed. Therefore, the recommendations in this section mittee suspects that the sensitivity is moderate to high and the
are based upon data that describe how accurate a diagnostic test specificity is poor if interpreted alone, but substantially better
is at confirming or excluding extrapulmonary TB, coupled with if interpreted in the context of the clinical setting, radiographic
evidence that the diagnosis of extrapulmonary TB leads to therapy findings, and other laboratory results. Most importantly, it
that improves patient-important outcomes. Tests used to diagnose is believed that cell counts and chemistries can provide use-
extrapulmonary TB are described in Supplementary Figure 1. ful information to guide the clinician toward either confirm-
With respect to the evidence that the diagnosis of extrapul- atory diagnostic testing for tuberculosis or diagnostic testing
monary TB leads to therapy that improves patient-important for alternative etiologies; this alone provides enough benefit
outcomes, trials directly comparing treatment with no treat- to justify the costs of the additional tests. The strength of the
ment will never be done. However, indirect evidence from recommendation is conditional because it is believed that the
patients with pulmonary TB (described above) and evidence balance of the benefits of the additional information versus the
from patients with disseminated TB suggests that extrapulmo- cost of the testing may be finely balanced in some clinical situ-
nary TB is treatable with high cure rates in most drug-suscepti- ations, and the quality of evidence provides little confidence in
ble cases and that untreated extrapulmonary TB has significant the estimates upon which the committee based its judgments.
morbidity and mortality, particularly meningeal TB [191]. As
an example, an observational study that excluded patients with Question 15: Should adenosine deaminase (ADA) and free
tuberculous meningitis found that mortality due to dissemi- IFN-γ levels be measured on specimens collected from sites
nated TB fell from 100% to <5% with the introduction of iso- of suspected extrapulmonary TB?
niazid-based antimicrobial regimens [192]. The large mortality Evidence
reduction in this study would probably have been less dramatic Test characteristics of ADA in the diagnostic evaluation of
if patients with tuberculous meningitis had been included, as meningeal, pleural, peritoneal, and pericardial tuberculosis

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e23


have been reported in meta-analyses of accuracy studies. Two risk of bias) and the studies reported variable results (ie, incon-
meta-analyses estimated the sensitivity and specificity of an ele- sistency), probably due in large part to the different thresholds
vated ADA level in the cerebrospinal fluid [195, 196]. The first used to define an elevated level.
meta-analysis included 10 studies and found a sensitivity and Recommendation 15a: We suggest that ADA levels be
specificity of 79% and 91%, respectively [195], using final clin- measured, rather than not measured, on fluid collected from
ical diagnosis, consistent pathology/cytology, or microbiologic patients with suspected pleural TB, TB meningitis, peritoneal
confirmation as the reference standard. Most of the studies used TB, or pericardial TB (conditional recommendation, low-qual-
a threshold of 9 U/L or 10 U/L to define an elevated ADA. The ity evidence).
second meta-analysis included 13 studies and showed that the Recommendation 15b: We suggest that free IFN-γ levels be
sensitivity and specificity are exquisitely sensitive to the thresh- measured, rather than not measured, on fluid collected from
old used to define an elevated ADA level [196]. If 4 U/L was patients with suspected pleural TB or peritoneal TB (condi-
used as the threshold, the sensitivity and specificity were >93% tional recommendation, low-quality evidence).
and <80%, respectively. In contrast, if 8 U/L was used as the
threshold, the sensitivity and specificity were <59% and >96%, Rationale
respectively. Neither the ADA level nor the IFN-γ level provide a defin-
Five meta-analyses that included 9–63 studies estimated that itive diagnosis of extrapulmonary TB disease; rather, they
the sensitivity and specificity of an elevated ADA level in the provide supportive evidence that must be interpreted in the
pleural fluid are 89%–99% and 88%–97%, respectively, with all entire clinical context. In any type of diagnostic testing for

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


but one of the meta-analyses estimating that the specificity is extrapulmonary TB, both false-negative results and false-pos-
≥90% [197–201]. A more recent meta-analysis reported similar itive results have important consequences. False-negative
sensitivity and specificity [202]. Final clinical diagnosis, con- results delay diagnosis and treatment while diagnostic testing
sistent pathology/cytology, or microbiologic confirmation were continues, whereas false-positive results may lead to unnec-
used as the reference standard in most studies. Thresholds used essary therapy and the associated risks of drug toxicity and
to define an elevated ADA level ranged from 10 U/L to 71 U/L, cost. Therefore, it is desirable for diagnostic tests to be both
with most clustering around 40 U/L. sensitive and specific.
A meta-analysis of 31 studies estimated that the sensitivity Our committee made the judgment that measurement of
and specificity of an elevated ADA level in pericardial fluid are ADA levels and free IFN-γ levels are indicated if false-neg-
88% and 83%, respectively [203]. The threshold to define an ele- ative results occur <30% of the time (ie, sensitivity is ≥70%)
vated ADA level was 40 U/L. Finally, a meta-analysis of 4 studies and false-positive results occur <20% of the time (ie, specificity
estimated that the sensitivity and specificity of an elevated ADA is ≥80%). The different thresholds for false results reflect the
level in peritoneal fluid are 100% and 97%, respectively [204]. committee’s recognition that the consequences of false-nega-
The threshold used to define an elevated ADA level ranged from tive results are generally transient, whereas the consequences
36 U/L to 40 U/L. of false-positive results may be long lasting. In this case, the
The test characteristics of free IFN-γ levels have not been as sensitivity and specificity of ADA were ≥79% and ≥83%,
extensively studied. A meta-analysis of 6 studies estimated that respectively, for detecting TB in cerebrospinal fluid, pleural
the sensitivity and specificity of an elevated free IFN-γ level fluid, peritoneal fluid, and pericardial fluid, so ADA measure-
in peritoneal fluid are 93% and 99%, respectively [205]. The ments are recommended in these fluids. Similarly, the sensitiv-
threshold used to define an elevated IFN-γ level ranged from ity and specificity of free IFN-γ measurements were ≥89% and
0.35 U/L to 9 U/L or 20 pg/mL to 112 pg/mL. A meta-analy- ≥97%, respectively, for detecting TB in pleural fluid and peri-
sis of 22 studies estimated that the sensitivity and specificity of toneal fluid, so free IFN-γ measurements are recommended in
an elevated free IFN-γ level in pleural fluid are 89% and 97%, these fluids.
respectively [206]. The threshold used to define an elevated The recommendations are conditional because the low qual-
IFN-γ level ranged from 0.3 U/L to 10 U/L or 12 pg/mL to ity of evidence does not provide sufficient confidence in the
300 pg/mL. We did not identify any studies that looked at the estimated sensitivities and specificities for the committee to
test characteristics of free IFN-γ levels on pericardial fluid or be certain that the balance of desirable to undesirable conse-
cerebrospinal fluid. quences favors testing and obtaining the specimens to test usu-
No studies were identified that reported the test character- ally requires an invasive procedure and, therefore, the balance
istics of using both ADA and free IFN-γ to evaluate specimens of benefits versus risks may vary substantially depending upon
from patients with suspected extrapulmonary TB. This evidence the clinical condition of the patient. Furthermore, the commit-
provides low confidence in the accuracy of the estimated test tee recognized that these tests often required the services of an
characteristics for both ADA and free IFN-γ levels. The studies off-site laboratory, and that standards were variable across labo-
did not report whether consecutive patients were enrolled (ie, ratories and across published studies.

e24 • CID 2017:64 (15 January)  •  Lewinsohn et al


Question 16: Should AFB smear microscopy be performed and obtaining the specimens to test usually requires an invasive
on specimens collected from sites of suspected extrapulmo- procedure and, therefore, the balance of benefits versus risks
nary TB? may vary substantially depending upon the clinical condition
Evidence of the patient.
The diagnostic yield and sensitivity of AFB smear micros-
copy tend to be lower in extrapulmonary TB than pulmo-
Question 17: Should mycobacterial cultures be performed on
nary TB because the former is paucibacillary. Accuracy
specimens collected from sites of suspected extrapulmonary
studies indicate that AFB smear microscopy has a sensitivity
TB?
of 0–10%, 14%–39%, 10%–30%, <5%, and 0–42% in pleural
Evidence
fluid (Supplementary Table 10), pleural tissue (Supplementary
Accuracy studies indicate that mycobacterial culture has a sen-
Table 10), urine (Supplementary Table 11), cerebrospinal fluid
sitivity of 23%–58%, 40%–58%, 80%–90%, 45%–70%, 45%–
(Supplementary Table  12), peritoneal fluid (Supplementary
69%, and 50%–65% in pleural fluid (Supplementary Table 10),
Table  13), and pericardial fluid (Supplementary Table  14),
pleural tissue (Supplementary Table 10), urine (Supplementary
respectively [207–218], when final clinical diagnosis, consistent
Table 11), cerebrospinal fluid (Supplementary Table 12), peri-
pathology/cytology, or microbiologic confirmation is used as
toneal fluid (Supplementary Table  13), and pericardial fluid
the reference standard. In contrast, the specificity of AFB smear
(Supplementary Table 14), respectively [207, 208, 211, 213–216,
microscopy tends to be quite high as described for pulmonary
219–223], when final clinical diagnosis, consistent pathology/
TB (≥90%). This evidence provides very low confidence in
cytology, or microbiologic confirmation is used as the reference

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


the estimated test characteristics because many studies do not
standard. The specificity of mycobacterial culture tends to be
report enrolling consecutive patients, most studies were small
comparatively higher than the sensitivity (>97%). This evi-
with few samples, and ranges are wide due to variable results
dence provides low confidence in the estimated test character-
from the individual studies.
istics because many studies do not report enrolling consecutive
Recommendation 16: We suggest that AFB smear micros-
patients and most studies were small with few samples.
copy be performed, rather than not performed, on specimens
Recommendation 17: We recommend that mycobacterial
collected from sites of suspected extrapulmonary TB (condi-
cultures be performed, rather than not performed, on spec-
tional recommendation, very low-quality evidence). Remarks:
imens collected from sites of suspected extrapulmonary TB
A positive result can be used as evidence of extrapulmonary
(strong recommendation, low-quality evidence). Remarks:
TB and guide decision making because false-positive results
A  positive result can be used as evidence of extrapulmonary
are unlikely. However, a negative result may not be used to
TB and guide decision making because false-positive results are
exclude TB because false-negative results are exceedingly
unlikely. However, a negative result may not be used to exclude
common.
TB because false-negative results are exceedingly common.

Rationale Rationale
AFB smear microscopy provides the opportunity for early diag- The estimated specificity of >97% for mycobacterial cultures in
nosis and treatment. The estimated specificity of ≥90% for AFB the diagnosis of extrapulmonary TB indicates that false-positive
smear in the diagnosis of extrapulmonary TB indicates that results occur only <3% of the time; thus, a positive mycobac-
false-positive results occur only ≤10% of the time; thus, if a pos- terial culture is a reliable indicator that infection is present. In
itive AFB smear result is obtained, it is reasonable to assume contrast, the estimated sensitivity of mycobacterial culture is
that infection is present and to act accordingly. In contrast, the widely variable depending upon the specimen type. However,
estimated sensitivity of <50% for all specimen types indicates even the specimen that provides the highest sensitivity (urine
that false-negative results are more common that true-nega- has an 80%–90% sensitivity for the diagnosis of urinary TB)
tive results and, thus, a negative AFB smear result should not provides false-negative results 10%–20% of the time. Such fre-
be used to exclude extrapulmonary TB; additional diagnostic quent false-negative results suggest that mycobacterial cultures
testing is indicated. should not be used exclude extrapulmonary TB; additional
Even though a positive AFB smear result is infrequent, the diagnostic testing is indicated.
committee judged the benefits of early diagnosis (early initia- The committee judged the diagnostic yield and benefits
tion of treatment, potential to reduce transmission) to outweigh of mycobacterial culture sufficient to outweigh the cost and
the cost and burden of AFB smear microscopy. The recommen- burden. Importantly, positive mycobacterial cultures are the
dation is conditional because the very low quality of evidence only way to obtain isolates for DST. Empiric treatment for
does not provide sufficient confidence in the estimated sensi- presumed drug-resistant TB may lead to unnecessary toxici-
tivities and specificities for the committee to be certain that the ties and cost if the patient actually has drug-susceptible TB,
balance of desirable to undesirable consequences favors testing whereas empiric treatment for drug-susceptible TB may lead

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e25


to treatment failure, risk of dissemination, and even death if mycobacterial culture results require at least 1–2 weeks,
the patient actually has drug-resistant TB. Moreover, delayed but NAAT can be performed within hours, thereby offer-
diagnosis of drug resistance will prolong therapy and increase ing the opportunity for early diagnosis and treatment. The
risk of default. committee felt that NAAT gives positive results frequently
The recommendation is strong despite the low quality of enough that the potential benefits outweigh the costs and
evidence because the committee is certain that the balance of burden of testing. Moreover, the committee felt that if the
desirable to undesirable consequences favors mycobacterial cul- test results are applied correctly (ie, a positive NAAT result
ture. This certainty reflects the committee’s recognition of the is considered adequate to confirm extrapulmonary TB, but
importance of obtaining mycobacterial isolates for DST com- a negative NAAT result is not used to exclude extrapulmo-
pared with the costs and burdens of performing the cultures, nary TB), then the risks associated with false results are
and the belief that additional data would not alter the balance minimal.
of desirable and undesirable consequences in the overwhelming The recommendation is conditional because the very low
majority of patients. quality of evidence does not provide sufficient confidence in
the estimated sensitivities and specificities for the committee to
be certain that the balance of desirable to undesirable conse-
Question 18: Should NAAT be performed on specimens col- quences favors testing and obtaining the specimens to test usu-
lected from sites of suspected extrapulmonary TB? ally requires an invasive procedure and, therefore, the balance
Evidence
of benefits versus risks may vary substantially depending upon
Meta-analyses have been published for the use of NAAT in

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


the clinical condition of the patient.
suspected pleural and meningeal tuberculosis [224, 225].
Most studies used final clinical diagnosis, consistent pathol- Cautions and Limitations
ogy/cytology, or microbiologic confirmation as the reference At this time there are no FDA-approved NAATs for use with
standard. Nucleic acid amplification performed on pleural extrapulmonary specimens.
fluid and cerebrospinal fluid has a sensitivity of 56% and 62%,
respectively, indicating false-negative rates of 44% and 38%, Question 19: Should histological examination be performed
respectively. In contrast, the specificity of NAAT is high for on specimens collected from sites of suspected extrapulmo-
both pleural fluid and cerebrospinal fluid (98% for both), indi- nary TB?
cating that only about 2% of positive results are false-positives. Evidence
Individual studies have been published describing the test char- Accuracy studies indicate that histological examination
acteristics of nucleic acid amplification on other body fluids has a sensitivity of 69%–97%, 86%–94%, 60%–70%, 79%–
and tissues. The studies showed considerable variability in the 100%, and 73%–100% in pleural tissue (Supplementary
sensitivity and specificity based upon the disease site; generally Table  10), urologic tissue (Supplementary Table  11), endo-
speaking, the sensitivity was usually <90%, while the specificity metrial curettage (Supplementary Table  11), peritoneal
was >95% (Supplementary Table 15) [191, 226]. This evidence biopsy (Supplementary Table  13), and pericardial tissue
provides very low confidence in the estimated test characteris- (Supplementary Table  14), respectively, when final clinical
tics because many studies do not report enrolling consecutive diagnosis, consistent pathology/cytology, or microbiologic
patients, findings were inconsistent as exemplified by the wide confirmation is used as the reference standard. The speci-
ranges, and the studies were small with few samples. ficity of mycobacterial culture microscopy tends to be low
Recommendation 18: We suggest that NAAT be performed, because necrotizing and nonnecrotizing granulomas are
rather than not performed, on specimens collected from sites seen in other infectious and noninfectious diseases. This
of suspected extrapulmonary TB (conditional recommendation, evidence provides very low confidence in the estimated test
very low-quality evidence). Remarks: A positive NAAT result can characteristics because many studies do not report enrolling
be used as evidence of extrapulmonary TB and guide decision consecutive patients, the wide ranges of sensitivity are due
making because false-positive results are unlikely. However, a to the variable results of individual studies, and the studies
negative NAAT result may not be used to exclude TB because were small with few samples.
false-negative results are exceedingly common. Recommendation 19: We suggest that histological examina-
tion be performed, rather than not performed, on specimens
Rationale collected from sites of suspected extrapulmonary TB (condi-
NAAT cannot replace mycobacterial culture for diagnosis tional recommendation, very low-quality evidence). Remarks:
because it is not sensitive enough and it does not produce Both positive and negative results should be interpreted in the
an isolate, which is needed for DST. However, NAAT is context of the clinical scenario because neither false-positive
appropriate as an adjunct to mycobacterial culture because nor false-negative results are rare.

e26 • CID 2017:64 (15 January)  •  Lewinsohn et al


Rationale Evidence
Tissue sampling with histological examination generally We identified no empirical evidence that estimated the fre-
occurs after other types of diagnostic testing have failed to quency with which the availability of genotyped isolates
identify a definitive diagnosis. Thus, at this stage in the diag- changed public health practices or affected patient outcomes.
nostic process, the committee thought testing was worthwhile Therefore, the recommendation is based upon the committee’s
if sensitivity and specificity were both >50%, meaning that collective clinical experience, which constitutes very low-qual-
true results were more likely than false results. Histological ity evidence.
examination surpassed these thresholds and, therefore, is rec- Recommendation 20: We recommend one culture isolate
ommended. However, the committee emphasizes the impor- from each mycobacterial culture–positive patient be submitted
tance of interpreting the results within the clinical context, to a regional genotyping laboratory for genotyping (strong rec-
to lessen the impact of false results. The recommendation is ommendation, very low-quality evidence).
conditional because the very low quality of evidence does not
provide sufficient confidence in the estimated sensitivities and Rationale
specificities for the committee to be certain that the balance Genotyping is useful in detecting false-positive results due to
of desirable to undesirable consequences favors testing and confirming laboratory cross-contamination [234, 235], investi-
obtaining the specimens to test usually requires an invasive gating outbreaks of TB (both detecting unsuspected outbreaks
procedure and, therefore, the balance of benefits versus risks and confirming suspected outbreaks) [236], evaluating contact
may vary substantially depending upon the clinical condition investigations [237], and determining whether new episodes

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


of the patient. of TB are due to reinfection or reactivation [238]. In addition,
genotyping is useful for elucidating sites and patterns of Mtb
GENOTYPING OF M. TUBERCULOSIS transmission within communities [237, 239]. This information
Over the past 2 decades, genotyping of TB strains has been shown is used by state and local tuberculosis control programs to focus
to be a valuable tool in TB control. Molecular epidemiology has interventions to interrupt further TB transmission. Genotyping
helped to identify unsuspected transmission, determine likely is used to aid public health departments in the control of TB
locations of transmission, measure the extent of transmission, and poses no risk to individual patients.
and differentiate reactivation from newly acquired infection Recently, whole-genome sequencing (WGS) has been applied
[227]. Often traditional contact investigations focus on persons to investigation of tuberculosis outbreaks [240]. This technique
in the household and workplace. Numerous reports describe TB may add discriminatory power to strain identification, but the
cases linked through genotyping of Mtb isolates, when detection role of WGS in outbreak investigation is still being determined.
of transmission was initially missed by conventional contact In response to nosocomial outbreaks and tuberculosis among
investigation because the setting was nontraditional. This type HIV-infected patients, the CDC established a national univer-
of transmission occurs frequently among members of a “social sal tuberculosis genotyping system for the United States. The
network” that is centered around a specific activity, including merger of modern molecular protocols for strain identification
illicit drug use, excess alcohol use, or gambling, or location such at the DNA level and conventional epidemiological methodol-
as a homeless shelter, adult entertainment club, or HIV residen- ogies has given birth to an enhanced collaborative strategy to
tial care facility [228–231]. When genotyping detects previously impact tuberculosis control efforts. Regional TB genotyping
unrecognized transmission of TB in a nonconventional setting, laboratories can be contacted through the state public health
public health interventions to contain and subsequently end the laboratories or TB control programs.
outbreak can be redirected to focus on the social network or The recommendation is strong because the committee felt
location associated with transmission. certain that the public health benefits of genotyping far outweigh
Genotyping or DNA fingerprinting of Mtb can be used for the modest costs and burdens of genotyping. Even though the
determining the clonality of bacterial cultures. PCR-based, evidence can provide very little confidence in the magnitude of
and sometimes Southern blotting, methods are used. The the benefits, costs, and burdens used by the committee to make
PCR-based methods are mycobacterial interspersed repeti- its decision, the differences seemed so overwhelming that the
tive units (MIRU) and spacer oligonucleotide typing (spol- committee thought it extraordinarily unlikely that additional
igotyping) [232, 233]. A  standardized protocol has been data would lead to a judgment that the costs and harms exceed
developed to permit comparison of genotypes from different the benefits.
laboratories [232].
RESEARCH NEEDS
Question 20: Should genotyping be performed on a culture As described by Abu-Raddad et  al [241], improved detection
isolate from culture-positive patients with TB? of those with TB and improved identification of those at risk

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e27


to progress once infected have the potential to substantially infected are at substantially lower risk, but remote infection
decrease the prevalence of TB and its associated mortality. remains difficult to define operationally. Those with nega-
tive TSTs or IGRAs are unlikely to progress to TB. However,
Tuberculosis the PPV of either test is relatively modest, as current estimates
The ability to rapidly and accurately identify Mtb as well as would suggest that among household contacts, 20–40 peo-
drug resistance (eg, through NAAT, line probe, molecular bea- ple require treatment to prevent one case. While the number
con, and Xpert MTB/RIF assays) reflects substantial advances. needed to treat based on IGRAs is not known with certainty,
While rapid tests for TB diagnosis still have a sensitivity of early evidence suggests that it is not likely to be dramatically
70%–90%, they may fail to detect paucibacillary pulmonary TB. different [24, 55]. Nonetheless, a careful evaluation of which
They also remain relatively expensive, making them difficult to diagnostic is more closely associated with the development of
implement in high-burden, low-resource settings. Ideally, what TB remains a research priority. Given the relatively poor PPVs
is needed is a simple, inexpensive, rapid (ie, hours) test that is of current diagnostics for the prediction of progression to TB
highly accurate (>95% sensitivity and specificity). Rapid tests disease, a diagnostic that can accurately identify those at risk
for detection of drug resistance are approaching the desired is needed. It should be noted that both QFT and T-SPOT are
level of accuracy, at least for rifampin. However, these tests also largely measures of CD4 T-cell immunity, but additional mark-
are relatively expensive and need to be expanded to allow for ers of inflammation, cellular, or humoral immunity may prove
detection of resistance to other TB medications. Such expansion useful. Clearly, the identification of biomarkers associated with
is currently limited by gaps in knowledge of the molecular basis the development of tuberculosis following infection will require

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


of resistance to most first- and second-line drugs. In this regard, carefully performed prospective investigation. In this regard,
improved functional tests for resistance may prove useful. the prospective evaluation of populations at risk for disease pro-
Other significant gaps remain in the diagnosis of pediatric gression, and the use of sophisticated imaging such as positron
and extrapulmonary TB. First, the yield of AFB smear and cul- emission tomography–computed tomography (PET-CT) are
ture in children is low compared to that in adults, which leads likely to further delineate markers associated with either disease
to excessive morbidity and mortality due to delayed and missed progression or subclinical infection.
diagnoses, especially in resource-limited settings. Conversely, Operationally, the intent of targeted testing is to identify those
the inability to exclude TB results in overtreatment when the who would benefit from treatment. While much is now known
diagnosis cannot be excluded. In areas of the world where TB is about the accuracy of both the TST and IGRAs, much less is
diagnosed entirely based on smear microscopy, children will be known about their performance with regard to treatment com-
almost completely neglected and untreated for TB. In areas with pletion. Additional research on the use of IGRAs with regard
greater resources, low yields of microbiologic specimens in chil- to provider and patient perceptions is needed to establish opti-
dren deter many clinicians from even attempting culture collec- mal diagnostic and treatment strategies. Finally, the literature
tion. This may result in prolonged treatment with extra TB drugs addressing the performance of IGRAs in children <5 years old
(in jurisdictions that use 4 drugs for 6 months in patients lack- is still limited and studies to inform the appropriate use of these
ing susceptibility data). Alternatively, drug resistance will not be tests to accurately diagnose LTBI in this age group are needed.
identified and the child could suffer dire consequences receiving Studies of young household contacts in low-incidence countries
inadequate care. Second, similar challenges exist for the accurate would be especially informative.
diagnosis of those with extrapulmonary TB. Finally, diagnostic
GUIDELINE STATEMENT
approaches to the identification of those likely to fail TB treat-
These guidelines are not intended to impose a standard of care. They provide
ment are needed. These limitations in the diagnosis of paucibac- the basis for rational decisions in the diagnostic evaluation of patients with
illary TB highlight the need to develop testing strategies based on possible latent tuberculosis or tuberculosis. Clinicians, patients, third-party
either host or bacterial markers of infection that can be measured payers, stakeholders, or the courts should never view the recommendations
contained in these guidelines as dictates. Guidelines cannot take into account
from readily available clinical sources such as plasma or urine. all of the often compelling unique individual clinical circumstances. Therefore,
no one charged with evaluating clinicians’ actions should attempt to apply
the recommendations from these guidelines by rote or in a blanket fashion.
Latent Tuberculosis Infection Qualifying remarks accompanying each recommendation are its integral parts
Individuals with immunological evidence of exposure to Mtb and serve to facilitate more accurate interpretation. They should never be
antigens, but without evidence of clinical disease are termed omitted when quoting or translating recommendations from these guidelines.
“latently” infected. However, it is clear that there is considerable
heterogeneity within this classification. As was described previ- Supplementary Data
ously, those with recent infection (<2 years) are at increased risk Supplementary materials are available at Clinical Infectious Diseases online.
Consisting of data provided by the author to benefit the reader, the posted
for progression to clinical disease, and functionally might be materials are not copyedited and are the sole responsibility of the author, so
considered acutely infected. Conversely, those more remotely questions or comments should be addressed to the author.

e28 • CID 2017:64 (15 January)  •  Lewinsohn et al


Notes 20. Marks SM, Taylor Z, Qualls NL, Shrestha-Kuwahara RJ, Wilce MA, Nguyen CH.
Outcomes of contact investigations of infectious tuberculosis patients. Am J
Acknowledgments.  The writing committee thanks Drs Mike Iseman
Respir Crit Care Med 2000; 162:2033–8.
and Jeffrey Starke for their critical examination of the manuscript. The
21. Reichler MR, Reves R, Bur S, et al. Contact Investigation Study Group. Treatment
committee is particularly indebted to Kevin Wilson for his patience and of latent tuberculosis infection in contacts of new tuberculosis cases in the United
his editing skills. States. South Med J 2002; 95:414–20.
Potential conflicts of interest.  D. L. C. has received speaking fees from 22. Hirsch-Moverman Y, Daftary A, Franks J, Colson PW. Adherence to treatment for
Qiagen. C. L. D. serves on the data and safety monitoring board (DSMB) for latent tuberculosis infection: systematic review of studies in the US and Canada.
Otsuka America Pharmaceutical, Inc, served on a DSMB for Sanofi Pasteur Int J Tuberc Lung Dis 2008; 12:1235–54.
Inc, received research support from Insmed, and received travel support 23. Huebner RE, Schein MF, Bass JB Jr. The tuberculin skin test. Clin Infect Dis 1993;
from Qiagen. L. R. serves as a speaker and on an advisory committee for 17:968–75.
24. Pai M, Denkinger CM, Kik SV, et al. Gamma interferon release assays for detec-
Boehringer Ingelheim and F.  Hoffmann-La Roche, serves on an advisory
tion of Mycobacterium tuberculosis infection. Clin Microbiol Rev 2014; 27:3–20.
committee and received research support from Biogen Inc, served on an
25. Kendig EL Jr, Kirkpatrick BV, Carter WH, Hill FA, Caldwell K, Entwistle M.
advisory committee for AstraZeneca, GlaxoSmithKline, and Sanofi Pasteur, Underreading of the tuberculin skin test reaction. Chest 1998; 113:1175–7.
served as a consultant for Bayer HealthCare, served as a speaker for Cipla. 26. Centers for Disease Control and Prevention. Guidelines for preventing the trans-
T.M.S. is employed by the US Centers for Disease Control and Prevention mission of Mycobacterium tuberculosis in health-care settings, 2005. MMWR
(CDC). T. R. S. serves on a DSMB for Otsuka. P. A. L. is employed by the Recomm Rep 2005; 54:1–141.
CDC. All other authors report no potential conflicts. All authors have sub- 27. Black CA. Delayed type hypersensitivity: current theories with an historic per-
mitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. spective. Dermatol Online J 1999; 5:7.
Conflicts that the editors consider relevant to the content of the manuscript 28. Cole ST, Brosch R, Parkhill J, et  al. Deciphering the biology of Mycobacterium
tuberculosis from the complete genome sequence. Nature 1998; 393:537–44.
have been disclosed.
29. Andersen P, Andersen AB, Sørensen AL, Nagai S. Recall of long-lived immunity
to Mycobacterium tuberculosis infection in mice. J Immunol 1995; 154:3359–72.
References 30. Mustafa AS, Amoudy HA, Wiker HG, et  al. Comparison of antigen-specific
1. Schünemann HJ, Osborne M, Moss J, et al. ATS Ethics and Conflict of Interest T-cell responses of tuberculosis patients using complex or single antigens of

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


Committee and the Documents Development and Implementation Committee. Mycobacterium tuberculosis. Scand J Immunol 1998; 48:535–43.
An official American Thoracic Society Policy statement: managing conflict of 31. Mustafa AS, Oftung F, Amoudy HA, et  al. Multiple epitopes from the
interest in professional societies. Am J Respir Crit Care Med 2009; 180:564–80. Mycobacterium tuberculosis ESAT-6 antigen are recognized by antigen-specific
2. Schunemann HJ, Oxman AD, Brozek J, et  al. Grading quality of evidence and human T cell lines. Clin Infect Dis 2000; 30(suppl 3):S201–5.
strength of recommendations for diagnostic tests and strategies. BMJ 2008; 32. Harboe M, Oettinger T, Wiker HG, Rosenkrands I, Andersen P. Evidence for
336:1106–10. occurrence of the ESAT-6 protein in Mycobacterium tuberculosis and virulent
3. Schünemann HJ, Jaeschke R, Cook DJ, et al. ATS Documents Development and Mycobacterium bovis and for its absence in Mycobacterium bovis BCG. Infect
Implementation Committee. An official ATS statement: grading the quality of Immun 1996; 64:16–22.
evidence and strength of recommendations in ATS guidelines and recommenda- 33. Mahairas GG, Sabo PJ, Hickey MJ, Singh DC, Stover CK. Molecular analysis of
tions. Am J Respir Crit Care Med 2006; 174:605–14. genetic differences between Mycobacterium bovis BCG and virulent M. bovis. J
4. World Health Organization. Global tuberculosis report 2015. Geneva, Bacteriol 1996; 178:1274–82.
Switzerland: WHO, 2015. 34. Berthet FX, Rasmussen PB, Rosenkrands I, Andersen P, Gicquel B. A
5. Edlin BR, Tokars JI, Grieco MH, et al. An outbreak of multidrug-resistant tuber- Mycobacterium tuberculosis operon encoding ESAT-6 and a novel low-mo-
culosis among hospitalized patients with the acquired immunodeficiency syn- lecular-mass culture filtrate protein (CFP-10). Microbiology 1998; 144 (pt
drome. N Engl J Med 1992; 326:1514–21. 11):3195–203.
6. Gandhi NR, Moll A, Sturm AW, et al. Extensively drug-resistant tuberculosis as a 35. Bothamley GH, Rudd RM. Clinical evaluation of a serological assay using a mon-
cause of death in patients co-infected with tuberculosis and HIV in a rural area of oclonal antibody (TB72) to the 38 kDa antigen of Mycobacterium tuberculosis. Eur
South Africa. Lancet 2006; 368:1575–80. Respir J 1994; 7:240–6.
7. Scott C, Kirking HL, Jeffries C, et al. Tuberculosis trends—United States, 2014. 36. Colangeli R, Spencer JS, Bifani P, et al. MTSA-10, the product of the Rv3874 gene
MMWR Morb Mortal Wkly Rep 2015; 64: 265–9. of Mycobacterium tuberculosis, elicits tuberculosis-specific, delayed-type hyper-
8. Bennett DE, Courval JM, Onorato I, et  al. Prevalence of tuberculosis infection sensitivity in guinea pigs. Infect Immun 2000; 68:990–3.
in the United States population: the national health and nutrition examination 37. Dillon DC, Alderson MR, Day CH, et al. Molecular and immunological charac-
survey, 1999–2000. Am J Respir Crit Care Med 2008; 177:348–55. terization of Mycobacterium tuberculosis CFP-10, an immunodiagnostic antigen
9. US Institute of Medicine; Committee on the Elimination of Tuberculosis in the missing in Mycobacterium bovis BCG. J Clin Microbiol 2000; 38:3285–90.
United States, Geiter L, ebrary Inc. Ending neglect the elimination of tuberculosis 38. Arend SM, de Haas P, Leyten E, et al. ESAT-6 and CFP-10 in clinical versus envi-
in the United States. Washington, DC: National Academy Press, 2000:xvi, 269 p. ronmental isolates of Mycobacterium kansasii. J Infect Dis 2005; 191:1301–10.
10. Reichman LB, Hershfield ES; ebrary Inc. Tuberculosis: a comprehensive inter- 39. Geluk A, van Meijgaarden KE, Franken KL, et al. Identification and characteriza-
national approach. Lung biology in health and disease v 144. 2nd ed. New York: tion of the ESAT-6 homologue of Mycobacterium leprae and T-cell cross-reactivity
Dekker, 2000:xxviii, 898 p. with Mycobacterium tuberculosis. Infect Immun 2002; 70:2544–8.
11. Targeted tuberculin testing and treatment of latent tuberculosis infection. 40. Geluk A, van Meijgaarden KE, Franken KL, et al. Immunological crossreactivity
American Thoracic Society. MMWR Recomm Rep 2000; 49:1–51. of the Mycobacterium leprae CFP-10 with its homologue in Mycobacterium tuber-
12. Reichler MR, Reves R, Bur S, et  al. Evaluation of investigations conducted to culosis. Scand J Immunol 2004; 59:66–70.
detect and prevent transmission of tuberculosis. JAMA 2002; 287:991–5. 41. Millington KA, Gooding S, Hinks TS, Reynolds DJ, Lalvani A. Mycobacterium
13. Catanzaro A. Nosocomial tuberculosis. Am Rev Respir Dis 1982; 125:559–62. tuberculosis-specific cellular immune profiles suggest bacillary persistence
14. Schluger NW, Rom WN. The host immune response to tuberculosis. Am J Respir decades after spontaneous cure in untreated tuberculosis. J Infect Dis 2010;
Crit Care Med 1998; 157:679–91. 202:1685–9.
15. Balasubramanian V, Wiegeshaus EH, Taylor BT, Smith DW. Pathogenesis of 42. Hinks TS, Dosanjh DP, Innes JA, et  al. Frequencies of region of difference 1
tuberculosis: pathway to apical localization. Tuber Lung Dis 1994; 75:168–78. antigen-specific but not purified protein derivative-specific gamma interferon-­
16. Dannenberg AM Jr. Immune mechanisms in the pathogenesis of pulmonary secreting T cells correlate with the presence of tuberculosis disease but do not
tuberculosis. Rev Infect Dis 1989; 11 Suppl 2:S369–78. distinguish recent from remote latent infections. Infect Immun 2009; 77:5486–95.
17. Horsburgh CR Jr. Priorities for the treatment of latent tuberculosis infection in 43. Kik SV, Franken WP, Arend SM, et al. Interferon-gamma release assays in immi-
the United States. N Engl J Med 2004; 350:2060–7. grant contacts and effect of remote exposure to Mycobacterium tuberculosis. Int J
18. Selwyn PA, Hartel D, Lewis VA, et al. A prospective study of the risk of tuberculo- Tuberc Lung Dis 2009; 13:820–8.
sis among intravenous drug users with human immunodeficiency virus infection. 44. Lalvani A, Nagvenkar P, Udwadia Z, et  al. Enumeration of T cells specific for
N Engl J Med 1989; 320:545–50. RD1-encoded antigens suggests a high prevalence of latent Mycobacterium tuber-
19. Markowitz N, Hansen NI, Wilcosky TC, et al. Tuberculin and anergy testing in culosis infection in healthy urban Indians. J Infect Dis 2001; 183:469–77.
HIV-seropositive and HIV-seronegative persons. Pulmonary Complications of 45. T-SPOT.TB: an aid in the diagosis of tuberculosis infection; package insert for in
HIV Infection Study Group. Ann Intern Med 1993; 119:185–93. vitro diagnositic use. Abingdon, UK: Oxford Immunotec Ltd.

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e29


46. Diagnostic Standards and Classification of Tuberculosis in Adults and Children. 71. Perez J, Kupper H, Spencer-Green G. Impact of screening for latent TB before
This Official Statement of the American Thoracic Society and the Centers for initiating anti-TNF therapy in North America and Europe. Ann Rheum Dis 2005;
Disease Control and Prevention was Adopted by the ATS Board of Directors, July 64:265.
1999. This statement was endorsed by the Council of the Infectious Diseases Society 72. Bartu V, Havelkova M, Kopecka E. QuantiFERON-TB Gold in the diagnosis of
of America, September 1999. Am J Respir Crit Care Med 2000; 161:1376–95. active tuberculosis. J Int Med Res 2008; 36:434–7.
47. Ferrara G, Losi M, D’Amico R, et al. Use in routine clinical practice of two com- 73. Bosshard V, Roux-Lombard P, Perneger T, et  al. Do results of the T-SPOT.TB
mercial blood tests for diagnosis of infection with Mycobacterium tuberculosis: a interferon-gamma release assay change after treatment of tuberculosis? Respir
prospective study. Lancet 2006; 367:1328–34. Med 2009; 103:30–4.
48. Ferrara G, Losi M, Meacci M, et  al. Routine hospital use of a new commercial 74. Chee CB, Gan SH, Khinmar KW, et al. Comparison of sensitivities of two com-
whole blood interferon-gamma assay for the diagnosis of tuberculosis infection. mercial gamma interferon release assays for pulmonary tuberculosis. J Clin
Am J Respir Crit Care Med 2005; 172:631–5. Microbiol 2008; 46:1935–40.
49. Lee JY, Choi HJ, Park IN, et al. Comparison of two commercial interferon-gamma 75. Chou CH, Hsu HL, Lee LN, Hsueh PR, Luh KT. Comparison of 2 interfer-
assays for diagnosing Mycobacterium tuberculosis infection. Eur Respir J 2006; on-gamma assays and Roche Cobas Amplicor Mycobacterium tuberculosis assay
28:24–30. for rapid diagnosis of tuberculosis among patients with suspected tuberculosis in
50. van Zyl-Smit RN, Zwerling A, Dheda K, Pai M. Within-subject variability of Taiwan. J Microbiol Immunol Infect 2009; 42:251–7.
interferon-g assay results for tuberculosis and boosting effect of tuberculin skin 76. Domínguez J, Ruiz-Manzano J, De Souza-Galvão M, et  al. Comparison of two
testing: a systematic review. PLoS One 2009; 4:e8517. commercially available gamma interferon blood tests for immunodiagnosis of
51. van Zyl-Smit RN, Pai M, Peprah K, et al. Within-subject variability and boosting tuberculosis. Clin Vaccine Immunol 2008; 15:168–71.
of T-cell interferon-gamma responses after tuberculin skin testing. Am J Respir 77. Dosanjh DP, Hinks TS, Innes JA, et  al. Improved diagnostic evaluation of sus-
Crit Care Med 2009; 180:49–58. pected tuberculosis. Ann Intern Med 2008; 148:325–36.
52. Metcalfe JZ, Cattamanchi A, McCulloch CE, Lew JD, Ha NP, Graviss EA. Test 78. US Food and Drug Administration. Medical devices: T-SPOT-TB—P070006.
variability of the QuantiFERON-TB gold in-tube assay in clinical practice. Am J Silver Spring, MD: FDA, 2008.
Respir Crit Care Med 2013; 187:206–11. 79. Gerogianni I, Papala M, Klapsa D, Zinzaras E, Petinaki E, Gourgoulianis KI.
53. Richeldi L. An update on the diagnosis of tuberculosis infection. Am J Respir Crit Whole-blood interferon-gamma assay for the diagnosis of tuberculosis infection
Care Med 2006; 174:736–42. in an unselected Greek population. Respirology 2008; 13:270–4.
54. Leyten EM, Prins C, Bossink AW, et  al. Effect of tuberculin skin testing on a 80. Goletti D, Carrara S, Butera O, et  al. Accuracy of immunodiagnostic tests for

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


Mycobacterium tuberculosis-specific IFN-{gamma} assay. Eur Respir J 2007; 29:1212–6. active tuberculosis using single and combined results: a multicenter TBNET-
55. Zellweger JP, Sotgiu G, Block M, et al. TBNET. Risk assessment of tuberculosis in Study. PLoS One 2008; 3:e3417.
contacts by IFN-γ release assays. A Tuberculosis Network European Trials Group 81. Harada N, Higuchi K, Yoshiyama T, et al. Comparison of the sensitivity and spec-
Study. Am J Respir Crit Care Med 2015; 191:1176–84. ificity of two whole blood interferon-gamma assays for M. tuberculosis infection.
56. Connell TG, Curtis N, Ranganathan SC, Buttery JP. Performance of a whole blood J Infect 2008; 56:348–53.
interferon gamma assay for detecting latent infection with Mycobacterium tuber- 82. Higuchi K, Kawabe Y, Mitarai S, Yoshiyama T, Harada N, Mori T. Comparison
culosis in children. Thorax 2006; 61:616–20. of performance in two diagnostic methods for tuberculosis infection. Med
57. Richeldi L, Bergamini BM, Vaienti F. Prior tuberculin skin testing does not boost Microbiol Immunol 2009; 198:33–7.
QuantiFERON-TB results in paediatric contacts. Eur Respir J 2008; 32:524–5. 83. Jafari C, Ernst M, Diel R, et al. Rapid diagnosis of smear-negative tuberculosis by
58. Bergamini BM, Losi M, Vaienti F, et al. Performance of commercial blood tests bronchoalveolar lavage enzyme-linked immunospot. Am J Respir Crit Care Med
for the diagnosis of latent tuberculosis infection in children and adolescents. 2006; 174:1048–54.
Pediatrics 2009; 123:e419–24. 84. Janssens JP. Interferon-gamma release assay tests to rule out active tuberculosis.
59. Kampmann B, Whittaker E, Williams A, et al. Interferon-gamma release assays do Eur Respir J 2007; 30:183–4; author reply 184–5.
not identify more children with active tuberculosis than the tuberculin skin test. 85. Kang YA, Lee HW, Hwang SS, et  al. Usefulness of whole-blood interfer-
Eur Respir J 2009; 33:1374–82. on-gamma assay and interferon-gamma enzyme-linked immunospot assay in
60. Lighter J, Rigaud M, Eduardo R, Peng CH, Pollack H. Latent tuberculosis diagno- the diagnosis of active pulmonary tuberculosis. Chest 2007; 132:959–65.
sis in children by using the QuantiFERON-TB Gold In-Tube test. Pediatrics 2009; 86. Latorre I, De Souza-Galvão M, Ruiz-Manzano J, et al. Quantitative evaluation of
123:30–7. T-cell response after specific antigen stimulation in active and latent tuberculo-
61. Talati NJ, Seybold U, Humphrey B, et  al. Poor concordance between inter- sis infection in adults and children. Diagn Microbiol Infect Dis 2009; 65:236–46.
feron-gamma release assays and tuberculin skin tests in diagnosis of latent 87. Leung WL, Law KL, Leung VS, et al. Comparison of intracellular cytokine flow
tuberculosis infection among HIV-infected individuals. BMC Infect Dis 2009; cytometry and an enzyme immunoassay for evaluation of cellular immune
9:15. response to active tuberculosis. Clin Vaccine Immunol 2009; 16:344–51.
62. Efficacy of various durations of isoniazid preventive therapy for tuberculosis: five 88. Liao CH, Lai CC, Tan CK, et al. False-negative results by enzyme-linked immu-
years of follow-up in the IUAT trial. International Union Against Tuberculosis nospot assay for interferon-gamma among patients with culture-confirmed
Committee on Prophylaxis. Bull World Health Organ 1982; 60:555–64. tuberculosis. J Infect 2009; 59:421–3.
63. Ferebee SH, Mount FW. Tuberculosis morbidity in a controlled trial of the 89. Meier T, Eulenbruch HP, Wrighton-Smith P, Enders G, Regnath T. Sensitivity
prophylactic use of isoniazid among household contacts. Am Rev Respir Dis of a new commercial enzyme-linked immunospot assay (T SPOT-TB) for diag-
1962; 85:490–510. nosis of tuberculosis in clinical practice. Eur J Clin Microbiol Infect Dis 2005;
64. Comstock GW, Ferebee SH, Hammes LM. A controlled trial of community-wide 24:529–36.
isoniazid prophylaxis in Alaska. Am Rev Respir Dis 1967; 95:935–43. 90. Palazzo R, Spensieri F, Massari M, et al. Use of whole-blood samples in in-house
65. Ferebee SH, Mount FW, Murray FJ, Livesay VT. A controlled trial of isoniazid bulk and single-cell antigen-specific gamma interferon assays for surveillance of
prophylaxis in mental institutions. Am Rev Respir Dis 1963; 88:161–75. Mycobacterium tuberculosis infections. Clin Vaccine Immunol 2008; 15:327–37.
66. Pape JW, Jean SS, Ho JL, Hafner A, Johnson WD Jr. Effect of isoniazid prophy- 91. Park SY, Jeon K, Um SW, Kwon OJ, Kang ES, Koh WJ. Clinical utility of the
laxis on incidence of active tuberculosis and progression of HIV infection. Lancet QuantiFERON-TB Gold In-Tube test for the diagnosis of active pulmonary
1993; 342:268–72. tuberculosis. Scand J Infect Dis 2009; 41:818–22.
67. Whalen CC, Johnson JL, Okwera A, et  al. A trial of three regimens to prevent 92. Ruhwald M, Bodmer T, Maier C, et al. TBNET. Evaluating the potential of IP-10
tuberculosis in Ugandan adults infected with the human immunodeficiency and MCP-2 as biomarkers for the diagnosis of tuberculosis. Eur Respir J 2008;
virus. Uganda-Case Western Reserve University Research Collaboration. N Engl 32:1607–15.
J Med 1997; 337:801–8. 93. Tan CK, Lai CC, Liao CH, et al. Rapid diagnosis of active pulmonary tuberculo-
68. Wilkinson D, Squire SB, Garner P. Effect of preventive treatment for tuberculosis sis in the elderly using enzyme-linked immunospot assay for interferon-gamma.
in adults infected with HIV: systematic review of randomised placebo controlled J Am Geriatr Soc 2009; 57:2361–2.
trials. BMJ 1998; 317:625–9. 94. Wang JY, Chou CH, Lee LN, et al. Diagnosis of tuberculosis by an enzyme-linked
69. Martinson NA, Barnes GL, Moulton LH, et al. New regimens to prevent tubercu- immunospot assay for interferon-gamma. Emerg Infect Dis 2007; 13:553–8.
losis in adults with HIV infection. N Engl J Med 2011; 365:11–20. 95. Bienek DR, Chang CK. Evaluation of an interferon-gamma release assay,
70. Carmona L, Gomez-Reino JJ, Rodriguez-Valverde V, et  al. Effectiveness of T-SPOT.TB, in a population with a low prevalence of tuberculosis. Int J Tuberc
recommendations to prevent reactivation of latent tuberculosis infection in Lung Dis 2009; 13:1416–21.
patients treated with tumor necrosis factor antagonists. Arthritis Rheum 2005; 96. LoBue PA, Moser KS. Use of isoniazid for latent tuberculosis infection in a pub-
52:1766–72. lic health clinic. Am J Respir Crit Care Med 2003; 168:443–7.

e30 • CID 2017:64 (15 January)  •  Lewinsohn et al


97. Nolan CM, Goldberg SV, Buskin SE. Hepatotoxicity associated with isoniazid 121. Liebeschuetz S, Bamber S, Ewer K, Deeks J, Pathan AA, Lalvani A. Diagnosis of
preventive therapy: a 7-year survey from a public health tuberculosis clinic. tuberculosis in South African children with a T-cell-based assay: a prospective
JAMA 1999; 281:1014–8. cohort study. Lancet 2004; 364:2196–203.
98. Yee D, Valiquette C, Pelletier M, Parisien I, Rocher I, Menzies D. Incidence of 122. Sun L, Xiao J, Miao Q, et  al. Interferon gamma release assay in diagnosis of
serious side effects from first-line antituberculosis drugs among patients treated pediatric tuberculosis: a meta-analysis. FEMS Immunol Med Microbiol 2011;
for active tuberculosis. Am J Respir Crit Care Med 2003; 167:1472–7. 63:165–73.
99. Comstock GW, Livesay VT, Woolpert SF. The prognosis of a positive tuberculin 123. Tsiouris SJ, Austin J, Toro P, et al. Results of a tuberculosis-specific IFN-gamma
reaction in childhood and adolescence. Am J Epidemiol 1974; 99:131–8. assay in children at high risk for tuberculosis infection. Int J Tuberc Lung Dis
100. Marais BJ, Gie RP, Schaaf HS, et al. The natural history of childhood intra-tho- 2006; 10:939–41.
racic tuberculosis: a critical review of literature from the pre-chemotherapy era. 124. Taylor Z, Nolan CM, Blumberg HM; American Thoracic Society; Centers
Int J Tuberc Lung Dis 2004; 8:392–402. for Disease Control and Prevention; Infectious Diseases Society of America.
101. Connell T, Bar-Zeev N, Curtis N. Early detection of perinatal tuberculosis Controlling tuberculosis in the United States. Recommendations from the
using a whole blood interferon-gamma release assay. Clin Infect Dis 2006; American Thoracic Society, CDC, and the Infectious Diseases Society of
42:e82–5. America. MMWR Recomm Rep 2005; 54:1–81.
102. Detjen AK, Keil T, Roll S, et al. Interferon-gamma release assays improve the 125. Greenaway C, Menzies D, Fanning A, Grewal R, Yuan L, FitzGerald JM;
diagnosis of tuberculosis and nontuberculous mycobacterial disease in chil- Canadian Collaborative Group in Nosocomial Transmission of Tuberculosis.
dren in a country with a low incidence of tuberculosis. Clin Infect Dis 2007; Delay in diagnosis among hospitalized patients with active tuberculosis–­
45:322–8. predictors and outcomes. Am J Respir Crit Care Med 2002; 165:927–33.
103. Dogra S, Narang P, Mendiratta DK, et al. Comparison of a whole blood inter- 126. Golub JE, Bur S, Cronin WA, et al. Delayed tuberculosis diagnosis and tubercu-
feron-gamma assay with tuberculin skin testing for the detection of tuberculosis losis transmission. Int J Tuberc Lung Dis 2006; 10:24–30.
infection in hospitalized children in rural India. J Infect 2007; 54:267–76. 127. Mase SR, Ramsay A, Ng V, et al. Yield of serial sputum specimen examinations in
104. Connell TG, Ritz N, Paxton GA, Buttery JP, Curtis N, Ranganathan SC. A the diagnosis of pulmonary tuberculosis: a systematic review. Int J Tuberc Lung
three-way comparison of tuberculin skin testing, QuantiFERON-TB gold and Dis 2007; 11:485–95.
T-SPOT.TB in children. PLoS One 2008; 3:e2624. 128. Steingart KR, Ng V, Henry M, et  al. Sputum processing methods to improve
105. Nicol MP, Davies MA, Wood K, et  al. Comparison of T-SPOT.TB assay and the sensitivity of smear microscopy for tuberculosis: a systematic review. Lancet
tuberculin skin test for the evaluation of young children at high risk for tubercu- Infect Dis 2006; 6:664–74.

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


losis in a community setting. Pediatrics 2009; 123:38–43. 129. Steingart KR, Henry M, Ng V, et al. Fluorescence versus conventional sputum
106. Hesseling AC, Mandalakas AM, Kirchner HL, et  al. Highly discordant T cell smear microscopy for tuberculosis: a systematic review. Lancet Infect Dis 2006;
responses in individuals with recent exposure to household tuberculosis. Thorax 6:570–81.
2009; 64:840–6. 130. Gordin F, Slutkin G. The validity of acid-fast smears in the diagnosis of pulmo-
107. Nakaoka H, Lawson L, Squire SB, et al. Risk for tuberculosis among children. nary tuberculosis. Arch Pathol Lab Med 1990; 114:1025–7.
Emerg Infect Dis 2006; 12:1383–8. 131. Yajko DM, Nassos PS, Sanders CA, Madej JJ, Hadley WK. High predictive value
108. Higuchi K, Harada N, Mori T, Sekiya Y. Use of QuantiFERON-TB Gold to inves- of the acid-fast smear for Mycobacterium tuberculosis despite the high preva-
tigate tuberculosis contacts in a high school. Respirology 2007; 12:88–92. lence of Mycobacterium avium complex in respiratory specimens. Clin Infect Dis
109. Chee CB, Soh CH, Boudville IC, Chor SS, Wang YT. Interpretation of the tuber- 1994; 19:334–6.
culin skin test in Mycobacterium bovis BCG-vaccinated Singaporean schoolchil- 132. Cruciani M, Scarparo C, Malena M, Bosco O, Serpelloni G, Mengoli C. Meta-
dren. Am J Respir Crit Care Med 2001; 164:958–61. analysis of BACTEC MGIT 960 and BACTEC 460 TB, with or without solid
110. Radhakrishna S, Frieden TR, Subramani R; Tuberculosis Research Centre media, for detection of mycobacteria. J Clin Microbiol 2004; 42:2321–5.
(ICMR). Association of initial tuberculin sensitivity, age and sex with the inci- 133. Forbes B, Banaiee N, Beavis K, et  al. Laboratory detection and identification
dence of tuberculosis in south India: a 15-year follow-up. Int J Tuberc Lung Dis of mycobacteria; approved guideline. Wayne, PA: Clinical and Laboratory
2003; 7:1083–91. Standards Institute (CLSI publication no. M48A), 2008.
111. Mancuso JD, Tribble D, Mazurek GH, et al. Impact of targeted testing for latent 134. Woods G, Lin G, Desmond E. Susceptibility test methods: mycobacteria,
tuberculosis infection using commercially available diagnostics. Clin Infect Dis Nocardia, and other actinomycetes. Washington, DC: ASM Press, 2011.
2011; 53:234–44. 135. Greco S, Girardi E, Navarra A, Saltini C. Current evidence on diagnostic accu-
112. Clark SA, Martin SL, Pozniak A, et  al. Tuberculosis antigen-specific immune racy of commercially based nucleic acid amplification tests for the diagnosis of
responses can be detected using enzyme-linked immunospot technology in pulmonary tuberculosis. Thorax 2006; 61:783–90.
human immunodeficiency virus (HIV)-1 patients with advanced disease. Clin 136. Ling DI, Flores LL, Riley LW, Pai M. Commercial nucleic-acid amplification tests
Exp Immunol 2007; 150:238–44. for diagnosis of pulmonary tuberculosis in respiratory specimens: meta-analysis
113. Sauzullo I, Mengoni F, Lichtner M, et al. QuantiFERON-TB Gold and selected and meta-regression. PLoS One 2008; 3:e1536.
region of difference 1 peptide-based assays for the detection of Mycobacterium 137. Dinnes J, Deeks J, Kunst H, et  al. A systematic review of rapid diagnostic
tuberculosis infection in a cohort of patients enrolled with suspected tuberculo- tests for the detection of tuberculosis infection. Health Technol Assess 2007;
sis. Diagn Microbiol Infect Dis 2008; 62:395–401. 11:1–196.
114. Vincenti D, Carrara S, Butera O, et al. Response to region of difference 1 (RD1) 138. Updated guidelines for the use of nucleic acid amplification tests in the diagnosis
epitopes in human immunodeficiency virus (HIV)-infected individuals enrolled of tuberculosis. MMWR Morb Mortal Wkly Rep 2009; 58:7–10.
with suspected active tuberculosis: a pilot study. Clin Exp Immunol 2007; 139. Shingadia D, Novelli V. Diagnosis and treatment of tuberculosis in children.
150:91–8. Lancet Infect Dis 2003; 3:624–32.
115. Cattamanchi A, Smith R, Steingart KR, et al. Interferon-gamma release assays for 140. Moore DF, Guzman JA, Mikhail LT. Reduction in turnaround time for labora-
the diagnosis of latent tuberculosis infection in HIV-infected individuals: a sys- tory diagnosis of pulmonary tuberculosis by routine use of a nucleic acid ampli-
tematic review and meta-analysis. J Acquir Immune Defic Syndr 2011; 56:230–8. fication test. Diagn Microbiol Infect Dis 2005; 52:247–54.
116. Dorman SE, Belknap R, Graviss EA, et al. Tuberculosis Epidemiologic Studies 141. Campos M, Quartin A, Mendes E, et  al. Feasibility of shortening respiratory
Consortium. Interferon-γ release assays and tuberculin skin testing for diagno- isolation with a single sputum nucleic acid amplification test. Am J Respir Crit
sis of latent tuberculosis infection in healthcare workers in the United States. Am Care Med 2008; 178:300–5.
J Respir Crit Care Med 2014; 189:77–87. 142. Sloutsky A, Han LL, Werner BG. Practical strategies for performance optimi-
117. Lobue P, Menzies D. Treatment of latent tuberculosis infection: an update. zation of the enhanced Gen-Probe amplified Mycobacterium tuberculosis direct
Respirology 2010; 15:603–22. test. J Clin Microbiol 2004; 42:1547–51.
118. Pai M, Joshi R, Dogra S, et al. Serial testing of health care workers for tuber- 143. Flores LL, Pai M, Colford JM Jr, Riley LW. In-house nucleic acid amplification
culosis using interferon-gamma assay. Am J Respir Crit Care Med 2006; tests for the detection of Mycobacterium tuberculosis in sputum specimens:
174:349–55. meta-analysis and meta-regression. BMC Microbiol 2005; 5:55.
119. Daley CL, Reves RR, Beard MA, et al. A summary of meeting proceedings on 144. Ling DI, Zwerling AA, Pai M. GenoType MTBDR assays for the diagno-
addressing variability around the cut point in serial interferon-gamma release sis of multidrug-resistant tuberculosis: a meta-analysis. Eur Respir J 2008;
assay testing. Infect Control Hosp Epidemiol 2013; 34:625–30. 32:1165–74.
120. Okada K, Mao TE, Mori T, et al. Performance of an interferon-gamma release 145. Morgan M, Kalantri S, Flores L, Pai M. A commercial line probe assay for the
assay for diagnosing latent tuberculosis infection in children. Epidemiol Infect rapid detection of rifampicin resistance in Mycobacterium tuberculosis: a system-
2007:1–9. atic review and meta-analysis. BMC Infect Dis 2005; 5:62.

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e31


146. Boehme CC, Nabeta P, Hillemann D, et al. Rapid molecular detection of tuber- 170. Sun SJ, Bennett DE, Flood J, Loeffler AM, Kammerer S, Ellis BA. Identifying
culosis and rifampin resistance. N Engl J Med 2010; 363:1005–15. the sources of tuberculosis in young children: a multistate investigation. Emerg
147. Centers for Disease Control and Prevention. Availability of an assay for detect- Infect Dis 2002; 8:1216–23.
ing Mycobacterium tuberculosis, including rifampin-resistant strains, and con- 171. Marquez L, Feske ML, Teeter LD, Musser JM, Graviss EA. Pediatric tuberculosis:
siderations for its use—United States, 2013. MMWR Morb Mortal Wkly Rep the litmus test for tuberculosis control. Pediatr Infect Dis J 2012; 31:1144–7.
2013; 62:821–7. 172. Anderson C, Inhaber N, Menzies D. Comparison of sputum induction with fib-
148. Woods G, Brown-Elliott B, Conville P, et al. Susceptibility testing of mycobac- er-optic bronchoscopy in the diagnosis of tuberculosis. Am J Respir Crit Care
teria, nocardiae, and other aerobic actinomycetes; approved standard—2nd ed. Med 1995; 152(5 pt 1):1570–4.
Wayne, PA: Clinical and Laboratory Standards Institute (CLSI publication no. 173. Conde MB, Soares SL, Mello FC, et al. Comparison of sputum induction with
M24-A2), 2011. fiberoptic bronchoscopy in the diagnosis of tuberculosis: experience at an
149. Francis J.  Curry National Tuberculosis Center and California Department acquired immune deficiency syndrome reference center in Rio de Janeiro, Brazil.
of Public Health. Drug-resistant tuberculosis: a survival guide for clinicians. Am J Respir Crit Care Med 2000; 162:2238–40.
2nd ed. San Francisco, CA: Francis J.  Curry National Tuberculosis Center, 174. McWilliams T, Wells AU, Harrison AC, Lindstrom S, Cameron RJ, Foskin E.
2008. Induced sputum and bronchoscopy in the diagnosis of pulmonary tuberculosis.
150. Chan ED, Laurel V, Strand MJ, et  al. Treatment and outcome analysis of 205 Thorax 2002; 57:1010–4.
patients with multidrug-resistant tuberculosis. Am J Respir Crit Care Med 2004; 175. Saglam L, Akgun M, Aktas E. Usefulness of induced sputum and fibreoptic
169:1103–9. bronchoscopy specimens in the diagnosis of pulmonary tuberculosis. J Int Med
151. Narita M, Alonso P, Lauzardo M, Hollender ES, Pitchenik AE, Ashkin D. Res 2005; 33:260–5.
Treatment experience of multidrug-resistant tuberculosis in Florida, 1994–1997. 176. Brown M, Varia H, Bassett P, Davidson RN, Wall R, Pasvol G. Prospective study
Chest 2001; 120:343–8. of sputum induction, gastric washing, and bronchoalveolar lavage for the diag-
152. Lew W, Pai M, Oxlade O, Martin D, Menzies D. Initial drug resistance and tuber- nosis of pulmonary tuberculosis in patients who are unable to expectorate. Clin
culosis treatment outcomes: systematic review and meta-analysis. Ann Intern Infect Dis 2007; 44:1415–20.
Med 2008; 149:123–34. 177. Al Zahrani K, Al Jahdali H, Poirier L, René P, Menzies D. Yield of smear, culture
153. Traore H, Fissette K, Bastian I, Devleeschouwer M, Portaels F. Detection of and amplification tests from repeated sputum induction for the diagnosis of pul-
rifampicin resistance in Mycobacterium tuberculosis isolates from diverse coun- monary tuberculosis. Int J Tuberc Lung Dis 2001; 5:855–60.
tries by a commercial line probe assay as an initial indicator of multidrug resist- 178. Charoenratanakul S, Dejsomritrutai W, Chaiprasert A. Diagnostic role of fiber-

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


ance. Int J Tuberc Lung Dis 2000; 4:481–4. optic bronchoscopy in suspected smear negative pulmonary tuberculosis. Respir
154. Moore M, Onorato IM, McCray E, Castro KG. Trends in drug-resistant tubercu- Med 1995; 89:621–3.
losis in the United States, 1993–1996. JAMA 1997; 278:833–7. 179. Wongthim S, Udompanich V, Limthongkul S, Charoenlap P, Nuchprayoon C.
155. LoBue PA, Moser KS. Isoniazid- and rifampin-resistant tuberculosis in San Fiberoptic bronchoscopy in diagnosis of patients with suspected active pulmo-
Diego County, California, United States, 1993–2002. Int J Tuberc Lung Dis 2005; nary tuberculosis. J Med Assoc Thai 1989; 72:154–9.
9:501–6. 180. Chawla R, Pant K, Jaggi OP, Chandrashekhar S, Thukral SS. Fibreoptic bron-
156. Clark CM, Li J, Driver CR, Munsiff SS. Risk factors for drug-resistant tuberculo- choscopy in smear-negative pulmonary tuberculosis. Eur Respir J 1988; 1:804–6.
sis among non-US-born persons in New York City. Int J Tuberc Lung Dis 2005; 181. Willcox PA, Benatar SR, Potgieter PD. Use of the flexible fibreoptic broncho-
9:964–9. scope in diagnosis of sputum-negative pulmonary tuberculosis. Thorax 1982;
157. Granich RM, Oh P, Lewis B, Porco TC, Flood J.  Multidrug resistance among 37:598–601.
persons with tuberculosis in California, 1994–2003. JAMA 2005; 293:2732–9. 182. Danek SJ, Bower JS. Diagnosis of pulmonary tuberculosis by flexible fiberoptic
158. Weiner M, Benator D, Burman W, et  al. Tuberculosis Trials Consortium. bronchoscopy. Am Rev Respir Dis 1979; 119:677–9.
Association between acquired rifamycin resistance and the pharmacokinetics of 183. Miro AM, Gibilara E, Powell S, Kamholz SL. The role of fiberoptic bronchoscopy
rifabutin and isoniazid among patients with HIV and tuberculosis. Clin Infect for diagnosis of pulmonary tuberculosis in patients at risk for AIDS. Chest 1992;
Dis 2005; 40:1481–91. 101:1211–4.
159. Helb D, Jones M, Story E, et al. Rapid detection of Mycobacterium tuberculosis 184. Salzman SH, Schindel ML, Aranda CP, Smith RL, Lewis ML. The role of bron-
and rifampin resistance by use of on-demand, near-patient technology. J Clin choscopy in the diagnosis of pulmonary tuberculosis in patients at risk for HIV
Microbiol 2010; 48:229–37. infection. Chest 1992; 102:143–6.
160. Zhu RY, Zhang KX, Zhao MQ, et al. Use of visual loop-mediated isotheral ampli- 185. Wallace JM, Deutsch AL, Harrell JH, Moser KM. Bronchoscopy and transbron-
fication of rimM sequence for rapid detection of Mycobacterium tuberculosis and chial biopsy in evaluation of patients with suspected active tuberculosis. Am J
Mycobacterium bovis. J Microbiol Methods 2009; 78:339–43. Med 1981; 70:1189–94.
161. World Health Organization. Molecular line probe assays for rapid screening 186. Kennedy DJ, Lewis WP, Barnes PF. Yield of bronchoscopy for the diagnosis of
of patients at risk of multi-drug resistant tuberculosis (MDR-TB). Geneva, tuberculosis in patients with human immunodeficiency virus infection. Chest
Switzerland: WHO, 2008. 1992; 102:1040–4.
162. Marais BJ, Brittle W, Painczyk K, et al. Use of light-emitting diode fluorescence 187. Sarkar SK, Sharma GS, Gupta PR, Sharma RK. Fiberoptic bronchoscopy in the
microscopy to detect acid-fast bacilli in sputum. Clin Infect Dis 2008; 47:203–7. diagnosis of pulmonary tuberculosis. Tubercle 1980; 61:97–9.
163. Loeffler AM. Pediatric tuberculosis. Semin Respir Infect 2003; 18:272–91. 188. Schoch OD, Rieder P, Tueller C, et al. Diagnostic yield of sputum, induced spu-
164. Marais BJ, Gie RP, Schaaf HS, Beyers N, Donald PR, Starke JR. Childhood pul- tum, and bronchoscopy after radiologic tuberculosis screening. Am J Respir Crit
monary tuberculosis: old wisdom and new challenges. Am J Respir Crit Care Care Med 2007; 175:80–6.
Med 2006; 173:1078–90. 189. Willcox PA, Potgieter PD, Bateman ED, Benatar SR. Rapid diagnosis of sputum
165. Berggren Palme I, Gudetta B, Bruchfeld J, Eriksson M, Giesecke J. Detection negative miliary tuberculosis using the flexible fibreoptic bronchoscope. Thorax
of Mycobacterium tuberculosis in gastric aspirate and sputum collected from 1986; 41:681–4.
Ethiopian HIV-positive and HIV-negative children in a mixed in- and outpa- 190. Pant K, Chawla R, Mann PS, Jaggi OP. Fiberbronchoscopy in smear-negative
tient setting. Acta Paediatr 2004; 93:311–5. miliary tuberculosis. Chest 1989; 95:1151–2.
166. Hatherill M, Hawkridge T, Zar HJ, et al. Induced sputum or gastric lavage for 191. Cheng VC, Yew WW, Yuen KY. Molecular diagnostics in tuberculosis. Eur J Clin
community-based diagnosis of childhood pulmonary tuberculosis? Arch Dis Microbiol Infect Dis 2005; 24:711–20.
Child 2009; 94:195–201. 192. Falk A. U.S.  veterans administration-armed forces cooperative study on the
167. Abadco DL, Steiner P. Gastric lavage is better than bronchoalveolar lavage for chemotherapy of tuberculosis. 13. Tuberculous meningitis in adults, with special
isolation of Mycobacterium tuberculosis in childhood pulmonary tuberculosis. reference to survival, neurologic residuals, and work status. Am Rev Respir Dis
Pediatr Infect Dis J 1992; 11:735–8. 1965; 91:823–31.
168. Steiner M, Zimmerman R, Park BH, Shirali SR, Schmidt H. Primary tuberculo- 193. Maartens G, Willcox PA, Benatar SR. Miliary tuberculosis: rapid diagnosis,
sis in children. 2. Correlation of susceptibility patterns of M. tuberculosis isolated hematologic abnormalities, and outcome in 109 treated adults. Am J Med 1990;
from children with those isolated from source cases as an index of drug-resistant 89:291–6.
infection in a community. Am Rev Respir Dis 1968; 98:201–9. 194. Kim JH, Langston AA, Gallis HA. Miliary tuberculosis: epidemiology, clinical
169. Steiner P, Rao M, Mitchell M, Steiner M. Primary drug-resistant tuberculo- manifestations, diagnosis, and outcome. Rev Infect Dis 1990; 12:583–90.
sis in children. Correlation of drug-susceptibility patterns of matched patient 195. Xu HB, Jiang RH, Li L, Sha W, Xiao HP. Diagnostic value of adenosine deam-
and source case strains of Mycobacterium tuberculosis. Am J Dis Child 1985; inase in cerebrospinal fluid for tuberculous meningitis: a meta-analysis. Int J
139:780–2. Tuberc Lung Dis 2010; 14:1382–7.

e32 • CID 2017:64 (15 January)  •  Lewinsohn et al


196. Tuon FF, Higashino HR, Lopes MI, et al. Adenosine deaminase and tuberculous 220. Seibert AF, Haynes J Jr, Middleton R, Bass JB Jr. Tuberculous pleural effusion.
meningitis–a systematic review with meta-analysis. Scand J Infect Dis 2010; Twenty-year experience. Chest 1991; 99:883–6.
42:198–207. 221. Simon HB, Weinstein AJ, Pasternak MS, Swartz MN, Kunz LJ. Genitourinary
197. Morisson P, Neves DD. Evaluation of adenosine deaminase in the diagno- tuberculosis. Clinical features in a general hospital population. Am J Med 1977;
sis of pleural tuberculosis: a Brazilian meta-analysis. J Bras Pneumol 2008; 63:410–20.
34:217–24. 222. Christensen WI. Genitourinary tuberculosis: review of 102 cases. Medicine
198. Liang QL, Shi HZ, Wang K, Qin SM, Qin XJ. Diagnostic accuracy of adeno- (Baltimore) 1974; 53:377–90.
sine deaminase in tuberculous pleurisy: a meta-analysis. Respir Med 2008; 223. Alvarez S, McCabe WR. Extrapulmonary tuberculosis revisited: a review of
102:744–54. experience at Boston City and other hospitals. Medicine (Baltimore) 1984;
199. Greco S, Girardi E, Masciangelo R, Capoccetta GB, Saltini C. Adenosine deami- 63:25–55.
nase and interferon gamma measurements for the diagnosis of tuberculous pleu- 224. Pai M, Flores LL, Hubbard A, Riley LW, Colford JM Jr. Nucleic acid amplifi-
risy: a meta-analysis. Int J Tuberc Lung Dis 2003; 7:777–86. cation tests in the diagnosis of tuberculous pleuritis: a systematic review and
200. Goto M, Noguchi Y, Koyama H, Hira K, Shimbo T, Fukui T. Diagnostic value of meta-analysis. BMC Infect Dis 2004; 4:6.
adenosine deaminase in tuberculous pleural effusion: a meta-analysis. Ann Clin 225. Pai M, Flores LL, Pai N, Hubbard A, Riley LW, Colford JM Jr. Diagnostic accu-
Biochem 2003; 40(pt 4):374–81. racy of nucleic acid amplification tests for tuberculous meningitis: a systematic
201. Ena J, Valls V, Pérez de Oteyza C, Enríquez de Salamanca R. The usefulness review and meta-analysis. Lancet Infect Dis 2003; 3:633–43.
and limitations of adenosine deaminase in the diagnosis of tubercular pleurisy. 226. Pai M, Ling DI. Rapid diagnosis of extrapulmonary tuberculosis using nucleic
A meta-analytical study [in Spanish]. Med Clin (Barc) 1990; 95:333–5. acid amplification tests: what is the evidence? Future Microbiol 2008; 3:1–4.
202. Gui X, Xiao H. Diagnosis of tuberculosis pleurisy with adenosine deaminase 227. Daley CL, Kawamura LM. The role of molecular epidemiology in contact inves-
(ADA): a systematic review and meta-analysis. Int J Clin Exp Med 2014; tigations: a US perspective. Int J Tuberc Lung Dis 2003; 7(12 suppl 3):S458–62.
7:3126–35. 228. Daley CL, Small PM, Schecter GF, et al. An outbreak of tuberculosis with accel-
203. Tuon FF, Litvoc MN, Lopes MI. Adenosine deaminase and tuberculous pericar- erated progression among persons infected with the human immunodeficiency
ditis—a systematic review with meta-analysis. Acta Trop 2006; 99:67–74. virus. An analysis using restriction-fragment-length polymorphisms. N Engl J
204. Riquelme A, Calvo M, Salech F, et al. Value of adenosine deaminase (ADA) in Med 1992; 326:231–5.
ascitic fluid for the diagnosis of tuberculous peritonitis: a meta-analysis. J Clin 229. Cluster of tuberculosis cases among exotic dancers and their close contacts—
Gastroenterol 2006; 40:705–10. Kansas, 1994–2000. MMWR Morb Mortal Wkly Rep 2001; 50:291–3.

Downloaded from http://cid.oxfordjournals.org/ by guest on January 10, 2017


205. Su SB, Qin SY, Guo XY, Luo W, Jiang HX. Assessment by meta-analysis of inter- 230. Bock NN, Mallory JP, Mobley N, DeVoe B, Taylor BB. Outbreak of tuberculosis
feron-gamma for the diagnosis of tuberculous peritonitis. World J Gastroenterol associated with a floating card game in the rural south: lessons for tuberculosis
2013; 19:1645–51. contact investigations. Clin Infect Dis 1998; 27:1221–6.
206. Jiang J, Shi HZ, Liang QL, Qin SM, Qin XJ. Diagnostic value of inter- 231. Increase in African immigrants and refugees with tuberculosis—Seattle-King
feron-gamma in tuberculous pleurisy: a metaanalysis. Chest 2007; County, Washington, 1998–2001. MMWR Morb Mortal Wkly Rep 2002;
131:1133–41. 51:882–3.
207. Epstein DM, Kline LR, Albelda SM, Miller WT. Tuberculous pleural effusions. 232. Cowan LS, Diem L, Monson T, et  al. Evaluation of a two-step approach for
Chest 1987; 91:106–9. large-scale, prospective genotyping of Mycobacterium tuberculosis isolates in the
208. Chan CH, Arnold M, Chan CY, Mak TW, Hoheisel GB. Clinical and patholog- United States. J Clin Microbiol 2005; 43:688–95.
ical features of tuberculous pleural effusion and its long-term consequences. 233. Oelemann MC, Diel R, Vatin V, et al. Assessment of an optimized mycobacte-
Respiration 1991; 58(3–4):171–5. rial interspersed repetitive-unit-variable-number tandem-repeat typing system
209. Berenguer J, Moreno S, Laguna F, et  al. Tuberculous meningitis in patients combined with spoligotyping for population-based molecular epidemiology
infected with the human immunodeficiency virus. N Engl J Med 1992; studies of tuberculosis. J Clin Microbiol 2007; 45:691–7.
326:668–72. 234. Breese PE, Burman WJ, Hildred M, et  al. The effect of changes in laboratory
210. Ogawa SK, Smith MA, Brennessel DJ, Lowy FD. Tuberculous meningitis in an practices on the rate of false-positive cultures for Mycobacterium tuberculosis.
urban medical center. Medicine (Baltimore) 1987; 66:317–26. Arch Pathol Lab Med 2001; 125:1213–6.
211. Haas EJ, Madhavan T, Quinn EL, Cox F, Fisher E, Burch K. Tuberculous menin- 235. Burman WJ, Reves RR. Review of false-positive cultures for Mycobacterium
gitis in an urban general hospital. Arch Intern Med 1977; 137:1518–21. tuberculosis and recommendations for avoiding unnecessary treatment. Clin
212. Monteyne P, Sindic CJ. The diagnosis of tuberculous meningitis. Acta Neurol Infect Dis 2000; 31:1390–5.
Belg 1995; 95:80–7. 236. Daley CL. Molecular epidemiology: a tool for understanding control of tubercu-
213. Karney WW, O’Donoghue JM, Ostrow JH, Holmes KK, Beaty HN. The spec- losis transmission. Clin Chest Med 2005; 26:217–31, vi.
trum of tuberculous peritonitis. Chest 1977; 72:310–5. 237. Malakmadze N, González IM, Oemig T, et al. Unsuspected recent transmission
214. Shakil AO, Korula J, Kanel GC, Murray NG, Reynolds TB. Diagnostic features of tuberculosis among high-risk groups: implications of universal tuberculosis
of tuberculous peritonitis in the absence and presence of chronic liver disease: a genotyping in its detection. Clin Infect Dis 2005; 40:366–73.
case control study. Am J Med 1996; 100:179–85. 238. Behr MA, Small PM. Molecular fingerprinting of Mycobacterium tuberculosis:
215. Sherman S, Rohwedder JJ, Ravikrishnan KP, Weg JG. Tuberculous enteritis how can it help the clinician? Clin Infect Dis 1997; 25:806–10.
and peritonitis. Report of 36 general hospital cases. Arch Intern Med 1980; 239. Kimerling ME, Benjamin WH, Lok KH, Curtis G, Dunlap NE. Restriction frag-
140:506–8. ment length polymorphism screening of Mycobacterium tuberculosis isolates:
216. Singh MM, Bhargava AN, Jain KP. Tuberculous peritonitis. An evaluation of population surveillance for targeting disease transmission in a community. Int J
pathogenetic mechanisms, diagnostic procedures and therapeutic measures. N Tuberc Lung Dis 1998; 2:655–62.
Engl J Med 1969; 281:1091–4. 240. Gardy JL, Johnston JC, Ho Sui SJ, et  al. Whole-genome sequencing and
217. Quale JM, Lipschik GY, Heurich AE. Management of tuberculous pericarditis. social-­network analysis of a tuberculosis outbreak. N Engl J Med 2011; 364:
Ann Thorac Surg 1987; 43:653–5. 730–9.
218. Fowler NO, Manitsas GT. Infectious pericarditis. Prog Cardiovasc Dis 1973; 241. Abu-Raddad LJ, Sabatelli L, Achterberg JT, et  al. Epidemiological benefits of
16:323–36. more-effective tuberculosis vaccines, drugs, and diagnostics. Proc Natl Acad Sci
219. Berger HW, Mejia E. Tuberculous pleurisy. Chest 1973; 63:88–92. U S A 2009; 106:13980–5.

Diagnosis of TB in Adults and Children  •  CID 2017:64 (15 January) • e33